Applications Of Field Theory Methods In Statistical Physics Of Nonequilibrium Systems 9789811229978, 2021005226, 2021005227, 9789811229985

Table of contents :
Contents
Preface
Introduction
1. Statistical Physics of Interacting Particle Systems
1.1 Systems of Particles with Interaction
1.2 Models of Statistical Physics
1.3 The Model of Hard Spheres with Attractive Interaction
1.4 Nonideal Gas at Low Temperatures
2. Statistical Description of Phase Transitions
2.1 Theory of the Second-Order Phase Transitions Bragg–Williams theory
2.2 Unification of the Theories of Phase Transitions
2.3 First-Order Phase Transitions
2.4 Dynamics of Metastable States
3. Path Integration and Field Theory
3.1 Classical and Quantum Systems
3.2 Saddle-Point Method or Stationary-Phase Method
3.3 Construction of the Field Theory
3.4 Hubbard–Stratonovich Transformation
3.5 The Mean-Field Theory and the Functional Integral
4. Peculiarity of Calculation of Some Models
4.1 Special Cases of the Calculation of Path Integrals
4.2 Harmonic Lattice Model
4.3 The n-Vector Model
4.4 Potts Model
4.5 Villain and Gauss Lattice Models
4.6 Two-Dimensional Coulomb-Gas Models
5. Statistical Description of Condensed Matter
5.1 Partition Function for Model Systems
5.2 Ideal Classical and Quantum Gases
5.3 Hard Spheres Model
5.4 Two Exactly Solvable Models of Statistical Physics
5.5 Gravitating Gas Model
5.6 Coulomb-like Systems
6. Inhomogeneous Distribution in Systems of Particles
6.1 Microcanonical Description of Gravitating Systems
6.2 Spatial Distribution Function
6.3 Inhomogeneity of Self-Gravitating Systems Statistical approach
6.4 Conditions for the Gravothermal Catastrophe Infinite system
6.5 Models with Attraction and Repulsion
7. Cellular Structures in Condensed Matter
7.1 Cellular Structures and Selection of States
7.2 Thermodynamic of Cellular Structures
7.3 Cellular Structures in Colloids
7.4 Geometry of the Distribution of Interacting Particles
8. Statistical Description of Nonequilibrium Systems
8.1 Nonequilibrium Gravitating Systems
8.2 Systems with Repulsive Interaction
8.3 Saddle States of Nonequilibrium Systems
8.4 Nonequilibrium Dynamics of Universe Formation
Conclusions
Bibliography
Index

Citation preview

Applications of Field Theory Methods in

Statistical Physics of Nonequilibrium Systems

12091 9789811229978 tp.indd 1

17/12/20 3:25 PM

b2530   International Strategic Relations and China’s National Security: World at the Crossroads

This page intentionally left blank

b2530_FM.indd 6

01-Sep-16 11:03:06 AM

Applications of Field Theory Methods in

Statistical Physics of Nonequilibrium Systems

Bohdan Lev National Academy of Science of Ukraine, Ukraine

Anatoly Zagorodny National Academy of Science of Ukraine, Ukraine

World Scientific NEW JERSEY

•

LONDON

12091 9789811229978 tp.indd 2

•

SINGAPORE

•

BEIJING

•

SHANGHAI

•

HONG KONG

•

TAIPEI

•

CHENNAI

•

TOKYO

17/12/20 3:25 PM

Published by World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, Singapore 596224 USA office: 27 Warren Street, Suite 401-402, Hackensack, NJ 07601 UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE

Library of Congress Cataloging-in-Publication Data Names: Lev, Bohdan, author. | Zagorodny, A., author. Title: Applications of field theory methods in statistical physics of nonequilibrium systems / Bohdan Lev, National Academy of Science of Ukraine, Ukraine, Anatoly Zagorodny, National Academy of Science of Ukraine, Ukraine. Description: New Jersey : World Scientific, [2021] | Includes bibliographical references and index. Identifiers: LCCN 2021005226 (print) | LCCN 2021005227 (ebook) | ISBN 9789811229978 (hardcover) | ISBN 9789811229985 (ebook) Subjects: LCSH: Statistical physics. | Field theory (Physics) Classification: LCC QC174.7 .L47 2021 (print) | LCC QC174.7 (ebook) | DDC 530.13/3--dc23 LC record available at https://lccn.loc.gov/2021005226 LC ebook record available at https://lccn.loc.gov/2021005227

British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library.

Copyright © 2021 by World Scientific Publishing Co. Pte. Ltd. All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the publisher.

For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy is not required from the publisher.

For any available supplementary material, please visit https://www.worldscientific.com/worldscibooks/10.1142/12091#t=suppl Typeset by Stallion Press Email: [email protected] Printed in Singapore

Lakshmi - 12091 - Applications of Field Theory Methods.indd 1

16/2/2021 10:02:23 am

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-fm

Preface In this book, we formulate a unified approach to the description of interacting many-particle systems combining the methods of statistical physics and quantum field theory. The strong point of this approach is that it may be applied to the description of phase transitions with the formation of new spatially inhomogeneous phases. Another advantage concerns the possibility to describe quasiequilibrium systems with spatially inhomogeneous particle distributions and metastable states. The validity of the methods of statistical description of manyparticle systems and models (theory of phase transitions included) is discussed and compared. The idea of using the quantum field theory approach and related topics (path integration, saddle-point and stationary-phase methods, Hubbard–Stratonovich transformation, mean-field theory, and functional integrals) is also described in sufficient detail for understanding as well as for further applications. To some extent, the book could be treated as a brief encyclopedia of methods applicable to the statistical description of spatially inhomogeneous equilibrium and metastable particle distributions. The additional strength of the book is that we not only formulate the general approach, but also apply it to solve various practical and important problems (gravitating gas, Coulomb-like systems, dusty plasmas, thermodynamics of cellular structures, nonuniform dynamics of gravitating systems, etc). The book could be useful not only for students, but also for researchers dealing with various methods of equilibrium and nonequilibrium statistical mechanics. v

page v

b2530   International Strategic Relations and China’s National Security: World at the Crossroads

This page intentionally left blank

b2530_FM.indd 6

01-Sep-16 11:03:06 AM

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-fm

page vii

Contents Preface

v

Introduction

1

1. Statistical Physics of Interacting Particle Systems 1.1 Systems of Particles with Interaction . . . . . 1.2 Models of Statistical Physics . . . . . . . . . . . 1.3 The Model of Hard Spheres with Attractive Interaction . . . . . . . . . . . . . . . . . . . . . . 1.4 Nonideal Gas at Low Temperatures . . . . . .

13 .... ....

13 20

.... ....

29 38

2. Statistical Description of Phase Transitions 2.1 Theory of the Second-Order Phase Transitions 2.2 Unification of the Theories of Phase Transitions . . . . . . . . . . . . . . . . . . . . . . . 2.3 First-Order Phase Transitions . . . . . . . . . . . 2.4 Dynamics of Metastable States . . . . . . . . . .

41 ...

41

... ... ...

46 61 73

3. Path Integration and Field Theory 3.1 Classical and Quantum Systems . . . . . . . . . . . . 3.2 Saddle-Point Method or Stationary-Phase Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Construction of the Field Theory . . . . . . . . . . . .

vii

79 79 88 93

February 5, 2021

viii

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-fm

page viii

Statistical Physics of Nonequilibrium Systems

3.4 Hubbard–Stratonovich Transformation . . . . . . . . 3.5 The Mean-Field Theory and the Functional Integral . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Peculiarity of Calculation of Some Models 4.1 Special Cases of the Calculation of Path Integrals . . . . . . . . . . . . . . . . . . . . . 4.2 Harmonic Lattice Model . . . . . . . . . . 4.3 The n-Vector Model . . . . . . . . . . . . . 4.4 Potts Model . . . . . . . . . . . . . . . . . . 4.5 Villain and Gauss Lattice Models . . . . 4.6 Two-Dimensional Coulomb-Gas Models

. . . . . .

. . . . . .

Partition Function for Model Systems . . . . Ideal Classical and Quantum Gases . . . . . Hard Spheres Model . . . . . . . . . . . . . . . Two Exactly Solvable Models of Statistical Physics . . . . . . . . . . . . . . . . . . . . . . . . 5.5 Gravitating Gas Model . . . . . . . . . . . . . 5.6 Coulomb-like Systems . . . . . . . . . . . . . .

98 102

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

5. Statistical Description of Condensed Matter 5.1 5.2 5.3 5.4

96

102 110 112 113 116 119 123

..... ..... .....

123 128 134

..... ..... .....

139 148 152

6. Inhomogeneous Distribution in Systems of Particles 6.1 Microcanonical Description of Gravitating Systems . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Spatial Distribution Function . . . . . . . . . . . 6.3 Inhomogeneity of Self-Gravitating Systems . . 6.4 Conditions for the Gravothermal Catastrophe 6.5 Models with Attraction and Repulsion . . . . .

163 . . . . .

. . . . .

. . . . .

7. Cellular Structures in Condensed Matter 7.1 Cellular Structures and Selection of States . . . . . . 7.2 Thermodynamic of Cellular Structures . . . . . . . .

163 174 184 189 209 216 216 218

February 5, 2021

Applications of Field Theory Methods. . . - 9in x 6in

14:3

b4116-fm

page ix

Contents

7.3 Cellular Structures in Colloids . . . . . . . . . . . . . . 7.4 Geometry of the Distribution of Interacting Particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8. Statistical Description of Nonequilibrium Systems 8.1 8.2 8.3 8.4

Nonequilibrium Gravitating Systems . . . Systems with Repulsive Interaction . . . . Saddle States of Nonequilibrium Systems Nonequilibrium Dynamics of Universe Formation . . . . . . . . . . . . . . . . . . . . .

ix

223 241

259 ...... ...... ......

259 274 282

......

295

Conclusions

305

Bibliography

313

Index

323

b2530   International Strategic Relations and China’s National Security: World at the Crossroads

This page intentionally left blank

b2530_FM.indd 6

01-Sep-16 11:03:06 AM

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Introduction The statistical description of interacting many-particle systems remains one of the key issues of theoretical physics. Such description is needed for understanding the thermodynamic, kinetic, electronic, and electromagnetic properties of a wide class of new substances that have been intensively studied recently (liquid crystals, dusty plasmas, low-dimensional structures in solids etc). Many traditional problems, e.g., phase transitions accompanied by the formation of structures and metastable spatially inhomogeneous states, as well as the description of stationary states in open systems, also require further studies. The main purpose of this book is to propose a new approach [1–6] to the statistical description of a system of interacting particles with regard for spatially inhomogeneous particle distributions. To describe such structures, it is necessary to work out a method that would enable us to select the states with thermodynamically stable particle distributions in the partition function. One such method makes use of the representation of the partition function in terms of a functional integral over auxiliary fields that makes it possible to employ the quantum field theory approach [7–14]. An attempt to apply the functional integral to the description of many-particles systems was discussed for the first time in [15]. Both advantages and challenges of this approach were described in [7]. Particularly, the advantage is that the extension of the functional integrals to the complex plane provides a possibility to apply the saddle-point method with no use of the perturbation theory. It allows to select

1

page 1

February 5, 2021

2

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Statistical Physics of Nonequilibrium Systems

and separate the system states associated both with homogeneous and inhomogeneous particle distributions. The description of the formation of spatially inhomogeneous particle and field distributions is of great importance for condensed matter physics. It concerns both the physical understanding of optimum states of the system and is valuable for applications in practice [16–18]. Earlier investigations of the formation conditions and behavior of inhomogeneous states have mainly employed the statistical theory of nonequilibrium processes. However, spatially inhomogeneous particle and field distributions can also be formed in equilibrium systems. The conditions for the formation of such structures and their physical manifestation are determined first of all by the type of interaction. So, we have to formulate an adequate mathematical method that would describe the formation and behavior of spatially inhomogeneous equilibrium particle and field distributions. A few model systems of interacting particles are known, for which the partition function can be evaluated exactly, at least in the thermodynamic limit. In this book, we demonstrate the efficiency of the proposed approach by a nonperturbed calculation of the partition function for the known model systems with interaction (hard spheres model, Coulomb gas, gravitating gas, etc.). This approach makes it possible to describe any system of interacting particles with regard to spatially inhomogeneous particle distributions. A typical physical situation that involves bound states in a particle system occurs when the interaction consists of long-range attraction and shortrange repulsion. Another realistic situation is associated with the opposite case when the repulsion range is longer than the attraction range. Such physical systems are, e.g., electrons on the liquid helium surface [19]; polar atoms and molecules on a metal or dielectric surface [20, 21]; ions implanted in silicon [18, 22]. If interaction of this type occurs, the system cannot be homogeneous and hence it involves spatially inhomogeneous particle distributions, namely, finite-size clusters. The proposed approach makes it possible to describe such particle distributions, to calculate cluster sizes, to estimate the number of

page 2

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Introduction

page 3

3

particles within a cluster, and to find the temperature of the phase transition to the state under consideration. The number of particles in a cluster and the size of the latter depend on the interaction type and its intensity as well as on external parameters. The residual interaction (uncompensated after the cluster formation) produces interaction between clusters that, in turn, can cause formation of new spatial structures in the cluster system. The problem of how to find the cluster distribution and to estimate the influence of the external factors also can be solved in terms of the proposed approach. Particularly, an approach to the statistical description of interacting particles and phase transitions accompanied by cluster formation was proposed in [3, 23]. Clusters are described by the function of spatial particle distribution. This function is a soliton solution of the nonlinear equation that arises in most cases of statistical description of interacting particles. The method developed in the papers [3, 23] is based on the application of the quantum field theory approach [8, 9, 13, 14, 24–26] and provides a possibility to find the spatial distribution of particles, to calculate the cluster size and to obtain the temperature of the phase transition to the state under consideration. By means of this description it has been shown that cluster formation is possible in the system of attracting particles. The basic equation for the function of spatial distribution in the hightemperature limiting case, i.e., in the case of Boltzmann statistics, has been found. However, the dependence of the equilibrium size of the cluster on the thermodynamic limit has not been determined, and the dynamics of its formation has not been considered. This approach is correct also for various statistics when the problem to describe the systems like gas of interacting Fermi and Bose particles arises, in which spatial inhomogeneity of this type can appear. Lately the interest on the Bose condensate of particles with negative scattering lengths has increased. The experiments [27–29] have shown that the condensate collapses when the number of particles reaches some critical value. The same result follows from the numerical solution of the Gross–Pitaevsky equation [30–32]. The collapse arises due to the tunneling through the barrier of particle attraction and quantum pressure that have given rise to the wave

February 5, 2021

4

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Statistical Physics of Nonequilibrium Systems

packet diffusion. Such situation can occur in the case of a shotrange attracting potential that can be described in terms of the scattering length. It is rather interesting to compare the stability of the model condensate with the long-range attracting potential, for example 1/R (gravitating gas), and the condensate with the short-range attracting potential (described by negative scattering length) by testing the solution of the Gross–Pitaevsky equation for stability. Perhaps, this model can be useful for investigating the early stages of the dynamically changing Universe [33]. Based on the statistical approach [3, 23], we describe the formation of a spatially inhomogeneous distribution in a system of interacting Bose particles. We obtain the conditions of cluster formation in the system for both Bose gas and condensate and describe the dynamics of cluster formation in the limiting case of high temperatures (Boltzmann statistics). We compare the properties of spatial inhomogeneity in the Bose condensate of particles with negative scattering lengths and particles with long-range attraction with respect to their instability to collapse. The statistical description of Coulomb-like systems is one of the key problems of statistical physics too. Such systems might be helpful in testing the ideas concerning the description of systems with long-range interactions in terms of statistical mechanics. Solving the problem under consideration is complicated by the fact that standard methods of statistical physics cannot be used in the case of a system with Coulomb interaction. The structure formation in a system of colloidal particles or in dusty plasmas provides typical examples of a system with Coulomb-like interaction. One of the ways to describe the spatially inhomogeneous distribution in a system of interacting particles is to use the new method. Many-particle systems with Coulomb interaction (Coulomb-like systems), such as plasmas, colloidal particles, electrolyte solutions, electron gas in solids, etc., are widely presented both in nature and under laboratory conditions. Many soft-matter systems, e.g., surfactant solutions, colloids in various solvents, and dust particles in plasmas, exhibit self-assembling into various structures. The interest in this system is generated by its applications to the studies of a

page 4

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Introduction

page 5

5

variety of peculiar phenomena in various fields of science [34–36]. One of the most important problems here is the statistical description of Coulomb-like systems with high concentrations of interacting particles [37]. The formation of various crystal structures, transitions between crystalline phases of different symmetries, and melting-like phenomena [3, 38] are observed when concentration increases [39]. Moreover, dusty plasmas as well as colloidal suspensions may be perfect media for the experimental investigation of classical fluids and solids [39–48] since both direct measurements of the interaction between colloidal particles [49, 50] and theoretical treatment [51] reveal the Coulomb-like nature of the interaction in colloidal suspensions [35]. It is rather difficult to solve this problem by traditional methods of statistical mechanics since they cannot be applied to inhomogeneous systems with Coulomb-like interactions. In such cases, specific methods should be used taking into account the inhomogeneity of particle distributions. In particular, these methods should employ an appropriate procedure to find the dominant contribution to the partition function and to avoid free-energy divergences when the volume grows infinite. Only a few model systems of interacting particles are known for which the partition function can be evaluated exactly in the thermodynamic limiting case [52–54] and only a few results describing equilibrium states have been obtained within the framework of equilibrium statistical mechanics. In our approach, we propose that the known results can be obtained much more easily in terms of the method of collective variables and integral transformations [55]. Moreover, this method makes it possible to obtain the free energy of a classical plasma system in a regular manner up to an arbitrary order. A universal sequence of ordered structures was obtained from the description of the self-assembly using functional and statistical field theory [56]. The most interesting and exciting problem in condensed matter physics is the study of first-order phase transitions that produce states accompanied by the formation of a new phase with nonzero order parameter [57–65]. Such spatially inhomogeneous states can be thermodynamically stable with respect to small perturbations.

February 5, 2021

6

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Statistical Physics of Nonequilibrium Systems

The relaxation towards a thermodynamically stable state is accompanied by the formation of critical nuclei of the new phase as the energy barrier that separates the nonequivalent minimum of the thermodynamic potential is overcome. All the characteristics of the first-order phase transition, i.e., the dimensions of the new phase bubbles, formation time, as well as time of relaxation towards a thermodynamically stable state, depend on the degree of this nonequivalence and the barrier [66, 69]. The required behavior of the free energy as a function of the order parameter can be realized either in the case of some special interparticle interaction in the system (in the microscopic treatment) [3, 7] or with the presence of an external field (in the phenomenological treatment) [67, 68, 70–76]. Interaction of the order parameter with the external fields of various nature [57–76] can modify the order of the transition and the intensity of such interaction determines the parameters of the new phase bubbles. In the general case, at both microscopic and phenomenological levels, the slowest subsystem is separated out while all features of the smaller-scale behavior are taken into account by averaging over probable fluctuations against the background of the order parameter selected. Choosing some certain order parameter facilitates the selection of probable states of the system at the microscopic level. From all probable microscopic states of the system, we select those that may be described in terms of the averaged order parameter whose variation scale is larger than the scale of probable fluctuations. The averaging results in the selection of probable states which may be described in terms of the selected order parameter associated with the “condensate” behavior of smallerscale fluctuations. At the phenomenological macroscopic level, the first-order phase transition accompanied with the new phase bubbles formation also selects probable fluctuations of the order parameter with condensate behavior of the wavelength shorter than the bubble size. The formation time and relaxation towards a thermodynamically stable state depend first of all on the fluctuations of the scale smaller than that of the order parameter inhomogeneity. Thus, when describing the first-order phase transitions, it is of crucial importance to separate out the slowest subsystem and to select the states that

page 6

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Introduction

page 7

7

contributes the most to the determination of the order parameter and the “condensate” correlated behavior of fluctuations. Only collective behavior of fluctuations of the scale smaller than the inhomogeneity of the characteristic variation of the order parameter can result in its destruction in the equilibrium case. In order to study a specific physical system we have to first determine the dynamical quantities that should describe the system and in some way characterize both the systems and its interaction with the environment. A physical system is usually surrounded by some other system, e.g., the whole Universe. The environmental system is sometimes referred to as a reservoir, and in some cases it is called a thermostat. The thermostat is actually the key factor for the thermodynamic parameters of the system, e.g., temperature, volume etc. Suppose the dynamical variables in an arbitrary point of space are determined by the variable that in the general case and in what follows is referred to as field. This variable is suitable to derive the expressions for thermodynamic quantities and to describe all probable changes occurring in the system. Such a field may imply, say, the occupation numbers, spin, dipole moment, as well as any other arbitrary quantity that describes specific properties of individual particles in an arbitrary point of the space. Moreover, the quantity under consideration is assumed to vary in space and to undergo all probable deviations from its average values (fluctuations). An important feature of the field is its geometric representation since it may be scalar, vector, or tensor of arbitrary rank. The geometric representation of the field is fairly important for the mathematical calculations of thermodynamic quantities and hence first of all we have to establish the physical characteristics of the system. In what follows, we restrict the analysis to the scalar field functions and consider the peculiarities of the calculation procedure for other representations where needed. We consider specific systems. On the other hand, phase transitions in nonequilibrium systems, accompanied by the formation of dissipative structures, are strongly influenced by external noises associated mainly with the fluctuations of the governing parameter (external field). The structure of the

February 5, 2021

8

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Statistical Physics of Nonequilibrium Systems

nonequilibrium thermodynamic potential is entirely similar to that of the equilibrium potential whose governing parameter is temperature. For a structure, i.e., a spatially inhomogeneous distribution of the order parameter, to be formed in the nonequilibrium case, again the symmetry of the thermodynamic potential as a function of the governing parameter must be violated [24, 66, 77, 78]. It appears that the interaction of the system with the external noise — governing parameter fluctuations — would provide the desired asymmetric behavior of the thermodynamic potential so as to have nonequivalent minimums for different values of the order parameter. In the case of noise-induced phase transitions, however, only the value of the governing parameter is changed, which is associated with the structure formation or phase transition. At the same time, the structure parameters depend, in the end, only on the degree of symmetry violation calculated in the traditional manner similar to the analysis of the thermodynamic potential nonlinearity [24–66, 68, 69]. In the case of nonequilibrium phase transitions, one can also trace the same approach as in the equilibrium description, with all probable fluctuations being averaged against the background of the separated order parameter. The separation of the order parameter is required for the experimental observation of the system and structural transformations occurring in it. The order parameter itself plays the role of a macroscopic parameter and the slowest variable in terms of which all the changes in the system can be described. However, a situation can occur, when the order parameter is not the slowest variable. An example of a nonequilibrium medium of this type is glass [79]. Equilibrium analogs of such systems exist as well. For example, the formation of a gas bubble in a superheated liquid, crystallization in a supercooled liquid, or superconductor in a magnetic field, for which the temperature or magnetic field can vary slower than the order parameter. The size of the second phase bubble depends on the observation temperature whereas the value averaged over various temperatures depends only on the character of interaction and on the dispersion of the governing parameter. Such cases are considered in our book.

page 8

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Introduction

page 9

9

The purpose of our study is to give a consistent description of phase transitions accompanied by the formation of new phase bubbles at both microscopic and phenomenological levels. Special attention is paid to the selection of states which contribute the most in the thermodynamic behavior of the system with inhomogeneous order parameter. Characteristic dimensions of the spatially inhomogeneous distribution of the order parameter may, in the end, be treated as a criterion for the selection of “condensed” fluctuations which make it possible to describe the system in terms of the order parameter. The book introduces the readers to a consistent description of spatially inhomogeneous systems. First of all, we describe the statistical approach to the study of interacting systems with probable formation of spatially inhomogeneous particle distributions. Formation of such spatially inhomogeneous structures may be regarded as a first-order phase transition since it implies the formation of a new phase, i.e., a finite-size cluster whose properties differ from those of the initial homogeneous state. In terms of statistical approach, probable reasons for inducing spatially inhomogeneous particle distributions are indicated and the principle is formulated for the selection of states that provide the required behavior of the free energy in terms of the macroscopic order parameter. Then the phenomenological approach is employed to show how the interaction of the separated order parameter with the fields of diverse nature can result in the change of the transition order; the parameters of the new phase formed are expressed in terms of the interaction intensity. Special attention is paid to the order parameter interaction with the field that varies slower than the order parameter. In this case, in a manner similar to the statistical analysis, the dimensions of the new phase of macrobubbles are expressed only in terms of the interaction parameters. Under the assumption that the field that is “external” with respect to the order parameter, fluctuates with proper dispersion, such noise-induced phase transition can be responsible for the formation of a new phase bubble with observed parameters.

February 5, 2021

10

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Statistical Physics of Nonequilibrium Systems

Finally, a probability description is proposed for a noise-induced first-order phase transition under the assumption of fluctuation variations of the governing parameter. These can be the variations of the field that is “external” with respect to the order parameter and can simulate local variations of the microscopic interaction energy or temperature. Thus a possibility is provided to analyze, in terms of the general approach, the effect of fluctuations of any governing parameter on the stationary macroscopic behavior of the system. A special emphasis is made, in all sections of the book, on the description of the state-selection methods that yield a correct phenomenological expression for the thermodynamic potential in terms of the observed order parameter. The self-gravitating systems and studying thereof have a fundamental physical background for testing the ideas concerning statistical and thermodynamic description of systems governed by long-range interaction. The statistical description of such systems is directly related to the problems of astrophysics [80, 81]. Some general problems at the field of self-gravitating systems have been studied for a long time [87, 88] and such systems seem to be very complicated when compared to other many-body systems. Normally, probable structure formation under various conditions can be described by the general methods. The notion of equilibrium for the systems mentioned above is not always well defined and such systems exhibit nontrivial behavior with the occurrence of phase transitions associated with the gravitational collapse. The standard methods of equilibrium statistical mechanics cannot be used to study self-gravitating systems given that the thermodynamic ensembles are not equivalent: negative specific heat [89] in the microcanonical ensemble does not exist in the canonical description [90]. Two types of approaches (statistical and thermodynamic) have been developed to determine the equilibrium states of self-gravitating systems and to describe probable phase transitions [90, 91]. It is generally believed that the mean-field theory is exact for self-gravitating systems. According to the article [92] and the book [52] one can come to the conclusion that the thermodynamic limit of a self-gravitating system does not exist. Nevertheless, it is

page 10

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Introduction

page 11

11

possible to take the usual thermodynamic limit and, consequently, to use safely the usual thermodynamic tools, primarily regularizing the long-range behavior of the gravitational potential and introducing a very large screening length. Only in this case, the system is thermodynamically unstable and the thermodynamic limit does exist [93]. Systems with long-range interactions, such as self-gravitating system, do not relax to the usual Gibbs thermodynamic equilibrium, but become trapped in quasi-stationary states whose lifetimes diverge as the number of particles grows. The theory that makes it possible to quantitatively predict the instability threshold of the spontaneous symmetry breaking for a class of d-dimensional self-gravitating systems was earlier presented in [94, 95]. Mostly due to the fact that self-gravitating systems exist in states far from equilibrium, the time of relaxation towards equilibrium state is very long. The homogeneous particle distribution in a self-gravitating system is not stable. The particle distribution in such a system initially is inhomogeneous in space. Therefore, the system brakes in a complex of inhomogeneous clusters that tend to collapse into a more condensed state. There were some attempts to include particle distribution inhomogeneity [3–5], but the solution has not yet been found. When the dependence of temperature on concentration is defined and, therefore, the concentration-dependence of pressure is defined too, one can obtain stable solutions for the gravitational formation of stars [90]. This approach seems to be somewhat inconsistent due to the fact that the equation of state should be obtained from the definition of the partition function though this definition for inhomogeneous systems is unknown [52]. Such inhomogeneous particle distribution, temperature, and chemical potential can be taken into account in the nonequilibrium statistical operator approach, where the possibility of a local change of thermodynamic parameters is considered [96]. The self-gravitating systems are nonequilibrium a priori . In the present book we propose a new approach based on a nonequilibrium statistical operator [96] that is more suitable for the description of gravitation systems. The equation of state and all

February 5, 2021

12

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-intro

Statistical Physics of Nonequilibrium Systems

needed thermodynamic characteristics are defined by the equations that govern the largest contributions to the partition function. Thus, there is no need to introduce an additional hypothesis about the density-dependence of the temperature. This dependence is obtained by solving relevant thermodynamic relations that describe the extreme nonequilibrium partition function. The probable spatially inhomogeneous distributions of particles and temperature are obtained for simple cases. In the equilibrium case, the well-known result [53, 54] for the partition function is reproduced. This approach is shown to be efficient to describe the inhomogeneous particle distributions and to find the thermodynamic parameters in a self-gravitating system. The main idea of this approach is to provide a detailed description of three-dimensional self-gravitating systems using the principles of nonequilibrium statistical mechanics and to obtain distributions of particles and temperature for fixed number of particles and energy within the system. We do not confine our inquiry to the dynamic aspects of this system, but describe possible inhomogeneous distributions of thermodynamic parameters for various external conditions. So, in the present book we formulate an adequate mathematical method that would describe the formation and behavior of spatially inhomogeneous distributions of particles and fields in both equilibrium and nonequilibrium cases.

page 12

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch01

page 13

Chapter 1

Statistical Physics of Interacting Particle Systems

1.1 Systems of Particles with Interaction The statistical and thermodynamic description of a system of interacting particles has been the basics for understanding condensed matter for many years. Significant success has been achieved as a result [80–82, 87, 88, 90–122]. Within the framework of this approach the phase transitions in a system of interacting particles from gas to liquid phases, and also from liquid to solid are described in [123–132]. Many models of the formation of separate phases in a system of interacting particles are considered depending on the value and character of the interparticle interaction [82, 97, 98]. Many systems may be treated as classical objects. These may be described by the classical Hamiltonian for N interacting particles, i.e., N   p2i + Vij , H= 2m i

(1.1)

i R0 * 0 where R0 = 3 3V 4π The solution (6.134) means that as long as two particles cannot come closer than their diameters, the system cannot be compressed to a volume smaller than V0 . Now we consider a system of particles interacting by gravitation attraction and hard sphere repulsion. For the Newtonian attraction the inverse operator is known as 1 Δr δrr , 4πGm2 where G is the gravitation constant, m is the particle mass, and Δr is the Laplace operator. Then, using the result of the hard sphere model (6.134), we may write the action as     V   1 V p2 (∇ϕ)2 1 3 ϕ dV + dV d p ln 1 − ξe exp −β S= 2β V0 4πGm2 ω V0 2m     V0  p2 1 dV d3 p ln 1 + ξ exp −β + ω 0 2m    V (∇ϕ)2 1 ϕ dV − 3 g5/2 (ξe ) + ln(1 − ξ) + (N + 1) ln ξ = 4rm λ V0 −1 Wrr  = −

+

V0 f (ξ) + ln(1 − ξ) + (N + 1) ln ξ, λ3 5/2

(6.135)

where rm = 2πGm2 β, and it should be noted that there appears the special Fermi function [53, 54]  ∞ ∞    4 ξl 2 dxx2 ln 1 + ξe−x = (−1)l+1 5/2 . f5/2 (ξ) = √ π 0 l l=1

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 201

Inhomogeneous Distribution in Systems of Particles

201

It occurs due to the repulsion of hard spheres that changes the behavior of Bose particles statistically. We introduce a dimensionless quantity r = R/rm instead of R, a 3 /λ3 . Given that the new variable σ = exp (ϕ/2), and denote α2 ≡ rm expression (6.135) contains a logarithm and 0 < ξ < 1, the condition imposed on the field σ is 0 < ξσ 2 < 1. Now, we consider the case ξ < 1 that corresponds to the nonoccurrence of the Bose condensate, i.e., no /V → 0. The action (in spherical coordinates) (6.135) may be written with a new abovementioned variable, i.e.,   ∞ +  , 1 ∂σ 2 − α2 g5/2 ξσ 2 r 2 dr S = 4π σ ∂r r0 +

V0 f (ξ) + (N + 1) ln ξ. λ3 5/2

(6.136)

The equation for the saddle point is the Lagrange equation for the function (6.136). Here it is given as       1 ∂σ 2 1 2 ∂g5/2 ξσ 2 2 ∂ 2 σ 1 ∂σ 2 ∂2σ σ ≡ − + α − ∂r 2 σ ∂r 2 ∂σ ∂r 2 σ ∂r ∞ l 2l−1 ξσ + α2 σ 2 = 0. (6.137) 3/2 l l=1 This equation has no analytical solution. We consider the limiting case ξ → 0 (Boltzmann gas). Then Eq. (6.137) is reduced to   1 ∂σ 2 ∂2σ − + ξα2 σ 3 = 0. (6.138) ∂r 2 σ ∂r This equation has a soliton solution [3] 1 Δ , σ=√ ξα cosh Δr

(6.139)

where Δ is an unknown integration constant that will be seen below. Any soliton solution corresponds to a spatially inhomogeneous distribution of particles — a finite-size cluster. The relevant asymptotics for (6.139) is σ 2 = 1 for r = d, where d is the cluster size, and σ → 0

February 5, 2021

202

Applications of Field Theory Methods. . . - 9in x 6in

14:3

b4116-ch06

page 202

Statistical Physics of Nonequilibrium Systems

as r → ∞. This solution describes the particles involved in the inhomogeneous formation of size d and the nonoccurrence of particles at infinity, since in this case the spatial distribution function is ρ (r) = mξ

1 2 σ λ3

(6.140)

3 with the normalization rm ρ (r) d3 r = mN . We substitute (6.139) in (6.136) bearing in mind that limT →∞ f5/2 (ξ) = ξ. Thus we have  S = 4π

d

r0

 V0 Δ2 − 2ξα2 σ 2 r 2 dr + 3 ξ + (N + 1) ln ξ. λ

(6.141)

Then we carry out integration using the expansion of 1/ cosh x ≈ 1 − x2 /2 in power series of x ≡ Δd 1, i.e., S = − (V − V0 )

Δ2 V0 + 3 ξ + (N + 1) ln ξ. 2 3 α λ λ

(6.142)

Here Δ2 is found from the asymptotics 1=

& Δ2 % 1 − Δ2 d2 =⇒ Δ2 ≈ ξα2 + ξ 2 d2 α4 . 2 ξα

Thus, assuming that V  V0 , we have S=−

V − V0 2 2 2 V − 2V0 ξ + (N + 1) ln ξ − ξ d α , 3 λ λ3

(6.143)

where ξ is found from the saddle point equation ∂S/∂ξ = 0 assuming that λ3 /V  λ6 /V 2 , ξ0  ξG , (ξ0 , and ξG are activities of the ideal sph are (5.28) and gravitating Bose gases, respectively, ξ0sph and ξG similar activities with the correction on the particle volume V0 ), i.e., ξ=

λ3 (N + 1) λ3 (N + 1) V02 λ3 (N + 1) 2d2 α2 λ6 (N + 1)2 + − ≈ 2 V − V0 V V2 (V − V0 ) " $ 2d2 α2 λ6 (N + 1)2 4d2 α2 λ6 (N + 1)2 V0 − + V2 V3

sph . = ξ0 + ξ0sph + ξG + ξG

(6.144)

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 203

Inhomogeneous Distribution in Systems of Particles

203

Then integrating (6.143) at the saddle point (6.144) in accordance with this formula, we obtain the partition function   2V V  0 sph 0 − 3 ξ −(N + 1) ZN = ZN × exp 3 ξ0sph + ξG + ξG λ λ "

sph ξ sph + ξG + ξG × ln 1+ 0 ξ0

$

 V − V0 2 2 2 ξ d α , × exp + λ3 

(6.145) 0 is the partition function of the ideal gas (5.32). The where ZN knowledge of the latter makes it possible to find the free energy of the system, i.e.,   2V V  sph 0 sph F = F0 − kT + ξ + ξ ξ − 3 ξ −(N + 1) G 0 G λ3 λ

"

sph ξ sph + ξG + ξG × ln 1+ 0 ξ0

$

 V − V0 2 2 2 ξ d α , − kT + λ3 

(6.146) where F0 is the free energy of the ideal Bose gas (5.33). Minimizing (6.146) by the cluster size d = D/rm and disregarding the correction sph , since on the volume V0 in the gravitation part of the activity ξG λ6 λ3 λ6 V0 ,

V3 V2 V we have dα2 λ3 2 (N + 1)2 ∂F = −kT ∂d V − V0



 d2 α2 λ3 (N + 1) 1−4 =0 V − V0 (6.147)

and obtain the optimum radius of the cluster to be   V − V0 V V0 2 = 1− d0 = 3 4N λ3 α2 4N rm V

(6.148)

February 5, 2021

204

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 204

Statistical Physics of Nonequilibrium Systems

or in the dimensionless variables D02

1 V kT = 4 2πGm2 N



 V0 1− . V

(6.149)

Now we consider the dynamics of cluster formation. To do this we employ the equation of motion given by ∂F ∂D = −χ , (6.150) ∂t ∂D where χ is the coefficient of reciprocal dimension of the diffusion mass flow through the cross-section. Making use of (6.147) and (6.149), we have  χN kT  ∂D = −D 3 + DD02 . (6.151) 4 ∂t 2D0 kT . Then we may rewrite Eq. (6.151) in a more We denote η ≡ χN 2D04 suitable form, i.e.,

D˙ + ηD 3 − D02 ηD = 0.

(6.152)

The solution of this equation under the condition that the initial state of the system is spatially homogeneous and the assumption that D(0)/D0 = 1/2 is given by D2 =

D02  . 1 + 3 exp −2ηD02 t

(6.153)

An analogous result (exponential approach to the equilibrium size) was obtained in paper [162]. Now let us consider the asymptotics of the solution (6.153) for stability. For this, we consider a small deviation from (6.153) X(t) = D(t) − D o (t), where

 0

D (t) =

0 D0

(6.154)

 at t → −∞ . at t → +∞

D 0 (t) = 0 means that the state of the system is spatially homogeneous, D 0 (t) = D0 means that the state is spatially inhomogeneous with

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 205

Inhomogeneous Distribution in Systems of Particles

205

the cluster of equilibrium size D0 . We denote B ≡ (ηD02 D − ηD 3 ). Expanding B(D) in power series of X(t) and neglecting powers higher than the first, we obtain an equation for the small deviation, i.e.,   dB X = D02 η − 3η(D 0 )2 X. X˙ = dD D=D0 The solution of this equation is given by   at D 0 = 0 exp(D02 ηt) . X(t) ∼ exp(−2D02 ηt) at D 0 = D0

(6.155)

(6.156)

Suppose the initial state of the system is spatially homogeneous (D = 0). It is not unstable because the small deviation (6.154) increases exponentially ∼ exp(kt) (where k is the Liapunov exponent in (6.156)). Some fluctuation of density that may occur in the system produces the gradient of the gravitation potential. Then the latter produces spatial inhomogeneity — a cluster with the size approaching the equilibrium value (6.149) asymptotically. This spatially inhomogeneous state with the cluster of size D = D0 is stable because the small deviation exponentially decreases, ∼ exp(−kt). Bose condensate Suppose we have two models of the Bose condensate. The first one suggests that particles interact by short-range attraction forces. Such interaction is described by the scattering length a < 0 [163]. The other condensate consists of hard spheres with the diameters asph > 0 interacting through the long-range gravitation forces proportional to 1/R. The gravitation interaction cannot be described by the scattering length because the change of the S-wave phase under scattering by the potential of effective radius r0 is given by the expression 1 1 r0 = − + k2 r0 + · · · a 2 while for the Newton potential r0 → ∞. Condensates with negative scattering lengths have been studied both in experimental [27–29] and theoretical [32, 58, 164–167] works.

February 5, 2021

206

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 206

Statistical Physics of Nonequilibrium Systems

It has been proved that such systems become unstable with collapse if the number of atoms attains the critical number Nc . Now we compare the properties of the Bose condensate of particles with negative scattering lengths and with long-range attraction to the instability towards collapse. It should be noticed that the Bose condensate is a continuous wave of matter — a coherent state [168], and hence it is described by some wave function that is the product of one-particle functions in the first approximation [58]. On the other hand, the apparatus of statistical mechanics is based on the postulate of random phases [53] that is not valid in the coherent state. Thus we cannot use the abovementioned method for the investigation of spatial inhomogeneity in the condensed phase. Hence, in order to study such model systems we apply the method based on the Gross–Pitaevsky equation [30, 31, 163] (we are considering the spherical-symmetry problem only). We have   2 ∂ 2 ψ 4π2 |a| 2 ∂ψ ∂ψ =− + N |ψ|2 ψ + V ψ, + i ∂t 2m ∂R2 R ∂R m (6.157)   4π2 asph 2 ∂ 2 ϕ 2 ∂ϕ ∂ϕ =− + N |ϕ|2 ϕ + U ϕ. + i ∂t 2m ∂R2 R ∂R m (6.158) Here ψ, ϕ are the wave functions of the condensates with densities ρ (R) = mN |ψ|2 or ρ (R) = mN |ϕ|2 ; m is the mass of a particle, and V is the energy of a particle in the external field (harmonic potential of the trap), i.e., V = mω 2 R2 /2.

(6.159)

U is the energy of a particle in the gravitation field of the condensate mass distributed by the law ρ (R)), i.e.,  ρ4πR2 dR . (6.160) U = −mG R This field produces a trap V due to the property of the long-range action.

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 207

Inhomogeneous Distribution in Systems of Particles

207

Equation (6.157) is a nonlinear differential equation of the second order with variable coefficients; Eq. (6.158) is a nonlinear integraldifferential equation of the second order with variable coefficients, Hence, we have to solve them numerically. These equations have soliton solutions under certain conditions that will be found in what follows. Similarly to the previous section, the availability of such solutions implies the occurrence of the spatial inhomogeneity of the system, i.e., a cluster. The stability of the soliton solution of Eq. (6.157) was investigated numerically in [32]. We shall study and compare some general properties of the stability of the soliton solution of Eqs. (6.157) and (6.158) based on the equation for the energy balance. We assume that the condensate may be characterized by the mean density ρ=

mN , V

where 4π V = 3

 3 L 2

is the volume of the system, L is the spatial region occupied by the condensate. Then the potential energy of the condensate as described by Eq. (6.157) is as follows W =

4π |a| 2 2 3 2 N N + mN ω 2 L2 , − 2 2mL mV 40

(6.161)

where the first addend is the energy of the quantum pressure [58] associated with the uncertainty principle. This energy is resistant to the compression of the gas. The second addend is the energy produced by the pseudo-potential [53]. This energy tends to compress the gas. The third addend is the energy of the condensate - L/2in the external field (6.159). This expression is obtained from 0 ρϕdV , where ϕ = ω 2 R2 /2 is the field potential of the trap, dV = 4πR2 dR is the differential volume.

February 5, 2021

208

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 208

Statistical Physics of Nonequilibrium Systems

We see that this Bose condensate is unstable. The instability towards collapse occurs due to tunnelling through the barrier of particle attraction and quantum pressure. Let us estimate the length and the height of the barrier. There is no need to find the exact values because the expression (6.161) is approximate and its derivation is an equation of the fifth order!!!. The length of the barrier is given by #  − |a| N. (6.162) l∼ mω The height of the barrier is given by Wm ∼

2 |a|2 N m

.

(6.163)

As follows from formula, (6.163), the barrier vanishes when the number of particles is greater than the critical number, . /mω . (6.164) Nc ∼ |a| Now, we estimate the region occupied by the gravitating Bose condensate and its energy. The potential energy of this condensate described by Eq. (6.158) is given by W =

4πasph 2 2 9 1 2 N N − G (mN )2 , + 2mL2 mV 10 L

(6.165)

where the first addend is the energy of quantum pressure [58]. The second addend is the energy produced by the pseudo-potential [53]. Given that asph > 0, this energy tends to expand the gas. The third addend is the energy in the gravitation -field of the L/2 condensate mass. - This expression is obtained from (1/2) 0 ρϕdV , where ϕ = −G (ρdV /R is the gravitation potential of the field of the condensate mass. This energy has a minimum W0G ∼ −

m5 N 3 G2 2

(6.166)

February 5, 2021

Applications of Field Theory Methods. . . - 9in x 6in

14:3

b4116-ch06

page 209

Inhomogeneous Distribution in Systems of Particles

209

at the point LG 0

2 ∼ 3 m NG

"

# 1+

$ N 2 asph Gm3 1+ . 2

(6.167)

We see that the gravitating Bose condensate is stable (it cannot collapse) unlike the condensate with the short-range attraction. Such behavior is caused by the long-range action of gravitation and the short-range attraction between particles. 6.5 Models with Attraction and Repulsion Now we consider a system of particles whose interaction consists of attraction and repulsion. This problem cannot be solved in the general case. Let us reveal the main features of spatially inhomogeneous particle distribution formation in cases when the inverse interaction operator is known. We consider the screened Coulomb repulsion and attraction. Making use of the known form of the inverse operators (5.12), we obtain   , 1 + (∇ϕ)2 + χ2 ϕ2 S = dV 2rm  + , 1 2 2 2 ϕ (∇ψ) + λ ψ − ξAe cos ψ + (N + 1) ln ξ, (6.168) + 2re 3/2  as before; χ−1 and λ−1 are the attraction where A ≡ 2πm/β2 and repulsion screening radii, respectively; rm ≡ 4πQ2 β, re ≡ 4πq 2 β; Q2 and q 2 are interaction constants. The saddle-point equations are given by  1  Δϕ − χ2 ϕ + ξAeϕ cos ψ = 0, rm  1  Δψ − λ2 ψ − ξAeϕ sin ψ = 0 re  dV ξAeϕ cos ψ = N + 1.

(6.169)

February 5, 2021

210

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 210

Statistical Physics of Nonequilibrium Systems

This set of nonlinear equations determines spatially inhomogeneous field distributions associated with the formation of finite-size clusters. In some cases, these equations can be solved analytically and thus the parameters of such formations can be found. Let us consider the case when the effective change of the parameters of the system occurs for distances λ−1 < r and χ = 0. Physically, this corresponds to the long-range attraction and shortrange repulsion. Suppose that ψ 1. We expand the second equation of the set (6.169) in power series of the “slow” field component ψ to obtain the relation ξAeϕ

λ2 ψ2 = 3 + 3ξAeϕ . 2 re

Having substituted the latter in (6.168), we may write the effective action as    1 3 λ2 3 2 3 ϕ ˜ (∇ϕ) + ξArm e + r + (N + 1) ln ξ, Seff = 2 dV 4 2 re m (6.170) where the dimensionless length r˜ = R/rm is introduced and the integration extends over the dimensionless volume V˜ . The physical situation described by the effective action (6.170) corresponds to the long-range gravitation attraction and effective repulsion for distances shorter than the interaction radius λ−1 . 3 and α2 ≡ 3λ2 r 3 /2r . Let us introduce the notation γ 2 ≡ ξArm e m Then we have the function σ = exp(ϕ/2) in the form   2  dσ 1 γ2 2 2 2 + γ σ + α + (N + 1) ln . Seff = 2 dV˜ 3 σ dr Arm (6.171) This function crucially differs from (5.82): the sign of the second term is opposite and, moreover, it contains an additional term giving rise to the interaction renormalization in the presence of effective repulsion. The extremum condition for (6.171) is realized in the

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 211

Inhomogeneous Distribution in Systems of Particles

211

solution of the equation 1 d2 σ − 2 dr σ



dσ dr

2

− γ 2σ3 = 0

(6.172)

with the first integral 

1 ∇σ σ

2

− γ 2 σ 2 = Δ2 .

(6.173)

The solution of the latter equation is given by σ ˜=

1 Δ , γ sinh Δ (r − r  )

where r  is the coordinate of the soliton center. As follows from the distribution function f (r) = Aξeϕ , this solution describes a spatially inhomogeneous particle distribution. We regard it as a finite-size cluster, with the cluster size to be determined. If a multi-soliton solution is realized, in which soliton centers are dispersed while numbers of particles in the solitons are 0 , where equal, then Seff = nSeff    % 2 & γ2 0 2 2 2 2 ˜ + k ln . (6.174) Seff = 8π r dr Δ + α + 2γ σ 3 Arm Here n is the number of clusters and k = (N + 1) /n is the number of particles within a cluster. In a manner similar to the analysis of new phase bubbles formation [24, 160], we write the effective action per cluster, i.e.,   2  1 ˜3  2 0 ˜ 2 S1 + k ln γ , R Δ + α2 + 2γ 2 R = 8π (6.175) Seff 3 3 Arm ˜ is the dimensionless cluster size and where R  R  1 σdσ . σ ˜ 2 dr = . S1 = 2 Δ + γ 2 σ2 2r0 σ0 In the latter expression, the cluster center is assumed to lie at the spherical coordinate system origin. Actually, S1 describes the cluster

February 5, 2021

212

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

Statistical Physics of Nonequilibrium Systems

surface energy. The above formulas are valid when the transition layer thickness is considerably smaller than the cluster size [33]. For physical reasons, the asymptotic behavior of the solution is R, then Δ ∼ R−1 and γe∼ Δ; for the following: σ 2 = 1 for r =  −1 −1 . Thus we obtain S1 − 2r0 γ 2 r = 2r0 , we have σ0 4γ 2 r02 and the effective action, in terms of the cluster size, is given by      ˜3  ˜2 R 1 R 0 2 3 ˜2 (6.176) Seff 8π α + R . − − k ln Arm ˜2 3 r0 R ˜ yields the value of Minimizing the action with respect to R ˜ R. It is evident that the solution with finite size of the spatially inhomogeneous particle distribution can be realized only for α2 Rr0 > 3. The phase transition occurs for R = 2r0 , this corresponds to the condition 2α2 r02 = 3. Thus, we obtain the value of the transition temperature as θc πb2 Q2 /r0 , where b = (Q/q) λr0  1. If in this case αR > 1, then the effective action reduces to   8π 2 ˜ 3 0 3 ˜2 α R − k ln Arm R . (6.177) Seff 3 ˜ we find the cluster Having minimized the action with respect to R, 3 2 ˜ = k/4πα , and the action to be given as size to be R 0     6 2 k 3/2 0 3 = k 1− − ln Arm . (6.178) Seff 2 ˜ r0 3 4πα α2 R 0 The minimum of this function is realized for the optimum value of the number of particles within a cluster that is determined by the equation " $ 4πα2 24π exp − . (6.179) kc = 1/3 3 )3/2 (Arm α4/3 kc r0 ˜ 3 = kc /4πα2 . Expanding the The critical size of a cluster is R c action (6.178) in power series in the vicinity of the critical value of

page 212

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 213

Inhomogeneous Distribution in Systems of Particles

213

the number of particles within a cluster, we obtain   2  k 2 1 0 = kc 1 − . Seff 3 2 kc 0 . The probability of finding a cluster of k particles is P ∼ exp −Seff Finally, the free energy of the gas of noninteracting clusters is F = −θ ln Z. In our case, with regard to the zero modes [178], this reduces to   1 6 0 0 (6.180) F = nθ Seff − ln Seff . 2 π

Model with long-range repulsion and short-range attraction Now let us consider the contrary case when the repulsion range is longer than the attraction range, so that λ = 0 and χ = 0. We assume that ϕ 1 and retain only the first term in the action expansion in power series of the fieldϕ. Then, we find from the first equation of the set (6.169) that ϕ˜ = ξArm /χ2 cos ψ and substitute this value in the action (6.168). Thus, we obtain    1 γ2 2 2 2 2 2 ˜ (∇ψ) − γ cos ψ − γ α cos ψ + (N + 1) ln , Seff = dV 2 Are3 (6.181) where we have introduced the dimensionless length r˜ = R/re and denoted γ 2 ≡ ξAre3 and α2 ≡ rm /2λ2 re3 . Since γα 1, we obtain an equation for ψ in the spherically symmetric case, i.e., d2 ψ 2 dψ − γ 2 sin ψ = 0. + dr 2 r dr

(6.182)

In the general case, the solution of this equation describes a soliton that may be regarded as a spatially inhomogeneous formation. Similarly to (5.83), we may disregard the second term when the dimension of this formation is larger than the transition layer

February 5, 2021

214

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

page 214

Statistical Physics of Nonequilibrium Systems

thickness [24, 160]. In this case, the first integral exists, i.e.,   1 dψ 2 + γ 2 cos ψ = C. (6.183) 2 dr For C = γ 2 , the solution of Eq. (6.182) is given by   ψ = arctan exp −γ r − r  .

(6.184)

Its asymptotics is ψ = π as r → 0, and ψ = 0 as r → ∞. Physically this solution describes the formation of a pore in a continuum distribution of particles, i.e., absence of particles within a limited volume of size d ∼ γ −1 that encloses the soliton center r  . In the case of a multi-soliton solution, which corresponds to the formation of a finite number of pores in the system, we may write       d   4γ 2 1 + γ 2 α2 4γ 4 α2 2 2 2 2 − −γ 1+γ α r dr Seff = 4πn cosh2 (yr) cosh4 (yr) 0   γ2 . − γ 2 1 + γ 2 α2 (V − Vd ) + (N + 1) ln Are3

(6.185)

Here we could use the results of the previous subsection and write the effective action in terms of the pore size. Our purpose, however, is to show the possibility of describing spatially inhomogeneous formations by statistical methods only. In order to obtain the final result we employ the integrals  d  d 5 r 2 dr r 2 dr ˜ = V = V˜d , and 4π 4π d 2 4 3 0 cosh (yr) 0 cosh (γr) where V˜d = (4/3)πd3 . Thus, we obtain the final expression for the action, i.e.,     γ2 2 2 2 ˜ 2 . Seff = 4γ 1 − γ α Vd − γ 2 1 + γ 2 α2 V˜ + (N + 1) ln 3 Are3 (6.186) The iteration procedure for γ 2 yields a simplified expression for the action, i.e.,   0 = −γ 2 V˜ + (N + 1) ln γ 2 /Are3 . Seff

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch06

Inhomogeneous Distribution in Systems of Particles

page 215

215

This corresponds to the approximation that all the pores occupy a volume that is small as compared to the system volume. Thus, we obtain γ˜2 = (N + 1) /V˜ . Substituting this result in the action (6.186), we obtain the final expression for the partition function of the system for the spatially inhomogeneous particle distribution, i.e.,    ˜ N +1 2 Vd 0 ZN = ZN . (6.187) 1 + γ 2 α2 − 4 1 − γ 2 α2 3 V˜ If we set γα 1 and Vd = V0 (the particle volume), then we obtain the same result as for the hard spheres model (5.50). Evidently, the phase transition occurs for γα → 1 that corresponds to the temperature θc 2πρQ2 /λ2 . The latter expression may be derived from the condition that the stability of the homogeneous distribution in the system of likely charged particles is violated.

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Chapter 7

Cellular Structures in Condensed Matter

7.1 Cellular Structures and Selection of States The resulting physical properties systems of interacting particles have been the subject of extensive studies [3, 8, 88]. The phase transitions and the ordering of a many-particle system as well as the properties of the latter are of fundamental importance in physics and for wide practical applications. The most interesting and challenging problem in condensed matter physics is the study of phase transitions with the formation of spatially inhomogeneous distributions of particles, clusters, or cellular structures [3]. These structures were observed in usual colloids [83, 179], in systems of particles introduced in liquid crystals [84, 180, 181], in the case of a ferromagnetic particle in a liquid crystal [85], and even in galactic systems [81]. In some systems, crystal-like ordering such as chain-like structures [182–184] and crystal structures was observed in liquid crystal colloids. As is evident from these examples, the formation of various structures in systems of interacting particles is not essentially dependent on the strength and character of the interaction, it can occur in various physical situations. This experimental result provides evidence that both determining the conditions and physical motivation of cellular structure formation indicate a very important problem in condensed matter physics. It is worth noting that the appearance of a cellular structure changes dramatically the properties of the condensed matter. Experimentally, it was found that, at the initial stage, a cellular structure originates 216

page 216

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

page 217

217

from a region with the lowest local density of particles. In all cases, the existing interaction leads to a spatially inhomogeneous distribution of particles with the formation of regions free from particles. The cellular structure should realize a minimum of free energy and provide stabilization of regions formed in a pure elastic medium, so that particles are placed at the boundaries of those regions. To describe the conditions for the formation and the properties of these particle structures, one should take into account all aspects of the particle interaction. For various physical cases, many different specialized explanations of the cellular structure formation have been proposed. Basically, the statistical description of many-particle systems concerns homogeneous states. In order to describe the behavior of a many-particle system with different strengths and features of the interaction, a method is required that would take into account the spatially inhomogeneous distribution of particles [3]. Such a method should consider the formation of a cluster, cellular structures, and structures with lower ordering. In this chapter, we present a general approach to describe cellular structures in various physical situations for different systems of interacting particles. The theoretical study of a colloidal system is usually focused either on the thermodynamic treatment or on general arguments. The essential point that is ignored by these arguments is that colloidal particles interact via their direct forces and indirect interaction mediated by the solvent phases. Actually, however, they are located within spatially nonuniform distributions of particles and hence interparticle interactions will also influence the physical behavior of the particle distribution and the aggregation process that governs the spatial distribution. This chapter presents an approach for the thermodynamic description of the cellular structure formation in usual colloids and liquid crystal colloids. It is not difficult to show that the cellular structure formation in such different systems is of similar nature. The formation of different structures depends on the initial concentration of particles and the character of interaction. The system tries to minimize the free energy by breaking it apart, and among many ways

February 5, 2021

218

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

to realize it, the system chooses nucleation and further formation of cellular structures. To characterize the aggregation effect is a very difficult task although many attempts have been made to measure fractal dimensional shapes of particle distributions. The method describes the experimental observations of similar systems with various types of interaction between particles. The general approach provides a possibility to estimate the cellular structure formation in various systems under different conditions. The characteristic of a structure (size a cluster or a cell) should depend only on the ratio of the interaction energy to temperature and concentration of particles. 7.2 Thermodynamic of Cellular Structures The theoretical description of systems of interacting particles is focused on the thermodynamic treatment or on the general arguments. First of all, we consider a system of noninteracting particles. Instead of investigating the nonuniform distribution of particles, we can discuss the probability of formation of voids without particles. To find voids or pores in the distribution of particles is not difficult. In the case of a continuous distribution of particles with concentration n = N/V , the size of a pore may be determined as the distance to the nearest particles. We can consider only the formation of spherical voids given that the shapes of voids are not of crucial importance and nonspherical shapes may be considered after a more precise study of various processes of void formation. In the case of a nonuniform distribution of noninteracting particles, the spherical shape of a void is approximated fairly well. To determine the probability of the void formation without particles we employ the Saslaw approach [81] to the gravitational system. The probability of the formation of voids depends on the distribution of nearest particles. For different points in space, the probability p(r)dr to find a particle at a distance between r and r + dr is equal to the probability of nonoccurrence of particles in the area with size r that should be multiplied by the probability to find a particle at distance r. This relation is given by    r   p(r )dr 4πnr 2 dr. (7.1) p(r)dr = 1 − o

page 218

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

The latter equation is equivalent to the one given by     p(r) p(r) d = −p(r) = − 4πnr 2 , dr 4πnr 2 4πnr 2

page 219

219

(7.2)

that has a solution   4πnr 2 p(r) = 4πnr 2 exp − . 3

(7.3)

The probability p(r)dr to find a particle at the distance between r and r + dr is equal to the probability of nonoccurrence of particles in the volume V multiplied by the probability ndV to find a particle within the volume between V and V + dV , i.e., p(r)dr = nP dV . From this relation, we obtain the probability of finding a void without particles with volume V as given by P (V ) = exp(−nV ) ≡ exp(−N ).

(7.4)

The latter expression has the form of Poisson distribution. It is obvious that in a system of noninteracting particles we can obtain only a nonuniform distribution of particles, but there exists a nonzero probability to find a void without particles with volume V . According to the fundamental principles of thermodynamics we can find an analytical result for the probability to find voids without particles by random selection in volume V in the case of a system of interacting particles. In this case we can employ the standard form of the grand canonical ensemble of N particles in volume V at temperature T [53] given by W (N ) = exp {βμN − βF } ,

(7.5)

where β = 1/kT is the reciprocal temperature, μ is the chemical potential, and F (N, V, T ) is the free energy that may be calculated by the canonical assemble. With this probability, we can find the averages of various physical quantities that depend on the number of particles. As an example, we employ the function z N [81] where z is an arbitrary variable. For the mean value of this function

February 5, 2021

220

Applications of Field Theory Methods. . . - 9in x 6in

14:3

b4116-ch07

page 220

Statistical Physics of Nonequilibrium Systems

we obtain   exp(N ln z)W (N ) = G−1 exp {(βμ + ln z)N − βF } z N  = N

N

≡ exp {Ψ(βμ + ln z) − Ψ(βμ)} .

(7.6)

The latter expression follows from the well-known relation between the partition function of the grand canonical ensemble G and the thermodynamic function, i.e., PV ≡ Ψ(μ, V, T ). kT

lnG(μ, V, T ) =

(7.7)

This is the thermodynamic definition of the average values of quantities that depend on the number of particles. We may also propose the statistical definition of the mean values of similar quantities that follows from the probability P (N ) to find N particles in volume V , i.e.,  z N P (N ), z N  = N

that should be equivalent to the thermodynamic expression       z N P (N ) = exp Ψ z(exp βμ) − Ψ exp(βμ) N

      ≡ exp −Ψ exp(βμ) exp Ψ z exp(βμ) .

(7.8)

Comparing equal number of terms of the power series in z on both sides of the equation thus obtained, we get P (N ) = exp(−Ψ) where (exp Ψ)N 0 =



exp(N βμ) (exp Ψ)N 0 , N!

d d(z exp(βμ))

N



 exp Ψ(z exp(βμ))

(7.9)

(7.10)

may be calculated for z = 0. Now, we can calculate the probability to find an empty void in the distribution of interacting particles.

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 221

Cellular Structures in Condensed Matter

This probability may be written as  PV . P (0) = exp − kT

221

(7.11)

If the equation of state is known, then in the general case, we can estimate the probability of voids existing without particles in the system of interacting particles. This relation shows that the existence of a void without particles depends on the equation of state for particles that fill these voids. It is given by a very good representative formula. Considering that the Laplace pressure of a void is P = 2σ/R, where R is the radius of a void, we obtain the probability in the well-known form  8πσR2 . (7.12) P (0) = exp − 3kT It is similar to the probability of the formation of a bubble of a new phase under the first-order phase transition. Thus we have a good argument to correct the description of the void formation. We should note that the formation of a cellular structure depends only on the thermodynamic properties of a system and can occur in any system of interacting particles and does not depend on the nature of interaction. Formation of a void produces the surface tension between the void and particles. Formation of a surface void should lead to the decrease of free energy of the system. This formula will be reconstructed by another approach in the next sections of this chapter. As we have already mentioned, if the equation of state is known, then we can estimate the probability of the existence of voids without particles for various systems of interacting particles. First of all, we consider the thermodynamic properties of a system with weak interaction between particles. In the general case, we can write the equation of state as PV = N (1 − nb2 ), (7.13) kT where b2 is the second virial coefficient. This is the Van der Waals equation of state where the second part represents the interaction in the system and may be obtained for various interesting situations.

February 5, 2021

222

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 222

Statistical Physics of Nonequilibrium Systems

For this system, the probability of a void in the distribution of interacting particles may be written as    PV ≡ exp −nV (1 − nb2 ) . (7.14) P (0) = exp − kT In the case with no interaction in the system, this formula reduces to the previous result of the Poisson distribution. The mean value of the size of a void without particles may be estimated by comparing the probability of the decrease e times. From this condition the estimated volume of a void is V = 1/[n(1 − nb2 )]. In the case of no interaction in the system, b2 = 0, this formula reduces to the previous result. In most cases of systems of interacting particles the virial coefficient b2 is negative and the volume of void is the smallest as in the case of noninteracting particles. In the case of a gravitating system of particles, the interaction energy has attractive nature and the virial coefficient is positive. This implies that the void volume becomes greater and thus in the gravitating system the empty area is greater. This result was obtained by the direct computer simulation for a system of equal particles with the gravitation Newtonian interaction. This result was obtained by the direct computer simulation for a system of particles with the gravitation Newtonian interaction (Fig. 7.1). For the next example, we consider the gas of hard spheres. We introduce the packing factor ν ≡ N V0 /V where V0 is the volume of one spherical particle. Then the equation of state of the gas of hard spheres may be written as V 1 + ν + ν2 − ν3 PV = ν . kT V0 (1 − ν)3

(7.15)

In this case, we compare the left part of this equation to one and thus estimate the volume of a void as V = V0

(1 − ν)3 . ν(1 + ν + ν 2 − ν 3 )

If ν → 1, then we obtain V → 0. In the case, of compact packing of hard spheres, it is obvious that voids cannot be formed. If ν → 0, then V = V0 (1/ν) and we reproduce the previous well-known result.

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

page 223

223

Fig. 7.1. The result of computer simulation for the system of particles with pure gravitation interaction [81].

7.3 Cellular Structures in Colloids For the past decade, colloidal systems have been used as model systems in an attempt to understand the phenomenon of the twodimensional phase transition. In all previous papers dealing with colloidal particles in fluids, attempts have been made to study the phase behavior near phase transitions in solvents with the formation of colloidal voids, soap froths, and clusters [83, 179, 194]. Most papers reported the occurrence of pattern formation by spherical particles trapped at the air–water interface, namely, two-dimensional structures. Depending on the size of a particle, the formation was observed of both reversible and irreversible clustering, that was due to the combination of electrostatic repulsive and long-range van der Waals attractive interactions. The formation of different structures depends on the initial concentration of particles, as one can easily see from the above formulas. The system tries to minimize the energy by breaking it apart, and out of the many ways of doing it, the system chooses the nucleation of voids and further formation of cellular

February 5, 2021

224

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

Fig. 7.2. The schematic presentation of the cluster or cellular structure formation in the system of interacting particles. The formation of structures depends on the concentration, interaction energy and temperature.

structures. The difference in probable situations can be seen in the schematic pictures (Fig. 7.2). A general description has been proposed related to the formation of various structures in a system of interacting particles. Basically a well-known approach is considered for the first-order phase transition. The analysis of this model enables one to predict a rich scenario of solvent-induced colloidal phase separation with the formation of clusters and cellular structures. In all cases, phase separation occurs in two inhomogeneous colloidal phases with different particle

page 224

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 225

Cellular Structures in Condensed Matter

225

densities [179]. If the interparticle interaction energy is known, then we can study the thermodynamic behavior of an aggregate of such particles and describe the conditions for the formation of a new structure. The character and intensity of the interparticle interaction in the system can produce a spatially inhomogeneous distribution of particles [3]. Phase transition is of the first-order when the external field imposes a surface boundary condition on all particles. Thus the scale is determined of the inhomogeneous distribution of colloidal particles. Usual colloids In order to demonstrate the mechanism of phase transition accompanied by the formation of an inhomogeneous distribution, we consider a system of spherical particles. In this case, the free energy of the solution of particles in a usual colloid in a self-consistent field in the many-body approximation may be written as F = Fp + Fs + Fn , where

 Fp =

U (r − r )f (r)f (r )drdr + · · ·

(7.16)

(7.17)

is the free energy in terms of the particle distribution function f (r) and U (r − r ) presents a sum over all interaction energies. The simplest free energy contains only the pair interaction potential. The entropy part of the free energy is given by the standard expression  (7.18) Fs = kT f (r) ln f (r) + [1 − f (r)] ln[1 − f (r)]dr. The entropy part of the free energy of this type causes the two classical particles not to be able to occupy their individual space positions. Next, the free energy resulting from the coupling between particles and the matter where these are immersed can be modeled as   W (ri − r)dr, (7.19) Fn = f (r) i

February 5, 2021

226

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

where W (r) contains microscopic information on the wetting properties of the surface of a particle located at the space point ri . In the case of liquid crystal colloids this part of the free energy may be presented as the anchoring energy. The minima of the free energy correspond to the self-consistent field solution for f (r). Each thermodynamic function of state corresponds to a solution that describes some phase of the particle arrangement. If their distribution is inhomogeneous, then the solution helps to find the stable phase associated with the nature of interaction and temperature. If the particle solution is disordered, then by definition the mean value f (ri ) = c, where c is the relative particle concentration. The concentration inhomogeneity gives rise to an additional term f (r) = c ±ϕ(r), where ϕ(r) describes the change of the probability distribution function of particles at different space points. If concentration inhomogeneities are smooth and their scale is much longer than the interparticle distance, then this quantity may be interpreted as the change of particle composition. When considering the continuum description, we can write the free energy increment associated with the inhomogeneous particle distribution by the power series expansion and using the long-wavelength expansion of the concentration, i.e. 1 ϕ(r ) = ϕ(r) + ρi ∂i ϕ(r) + ρi ρj ∂j ∂i ϕ(r). 2 Here ρ = r − r is the distance between particles. In the longwavelength approximation we can rewrite this part of the free energy in the standard form as    1 2 2 1 4 1 2 2 l (∇ϕ) − μ ϕ + λϕ − εϕ , (7.20) ΔF (ϕ) = dr 2 2 4   where μ2 ≡ (V − kT )/(c(1 − c)), V = U (ρ)dρ and l2 = U (ρ)ρ2 dρ is determined through the interaction energy, ρ being the distance between the particles. The coefficient λ is responsible for the nonlinearity of the system that is induced by the many-body interaction in the system of particles. The new coefficient ε = N 4πR02 W represents the energy that is involved in each particle through the wetting

page 226

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 227

Cellular Structures in Condensed Matter

227

effect. Here R0 is the particle size and W is the energy for anchoring molecules of the elastic medium to the surface of the particle. This coefficient may be introduced provided the wetting effect is similar for each particle. In this case, the general presentation of the free energy should not contain any even terms because the distribution function of macroparticles satisfies the relation f (r)dr = N ,  ϕ(r)dr = 0. This expression describes the Landau free energy of a system of particles that are foreign in the elastic medium below the phase-transition temperature of this system. Thus, we can see that the minimum of the free energy realizes a spatially inhomogeneous particle distribution only provided the sign satisfies some relation, and the values of coefficients are determined by the interparticle interaction [66]. In order to find the condition under which the homogeneous particles distribution becomes unstable, we should calculate all the coefficients. The temperature of the phase transition to new states can be determined from the relation kTc = c(1 − c)V [68]. This is the well-known function that describes the first-order phase transition accompanied by the cluster formation in a system of interacting particles [3, 66]. Thus, the description follows the concentration. The most important contribution to the concentration is associated with the field configuration for which the value of the free energy is minimum, i.e., Δϕ −

dΦ = 0, dϕ

(7.21)

where the potential is given by 1 1 Φ = − μ2 ϕ2 + λϕ4 − εϕ. 2 4

(7.22)

Substituting the solution of the previous equation in the expression for free energy yields the condition for the formation of a new phase. When the difference of the minima of the effective potential values is greater than the barrier height, then the free energy may be presented

February 5, 2021

228

Applications of Field Theory Methods. . . - 9in x 6in

14:3

b4116-ch07

page 228

Statistical Physics of Nonequilibrium Systems

in the form 

∞

ΔF = 2π 0



1 dϕ 2 4π r dr + Φ (ϕ) = − r 3 ε + 4πr 2 σ, 2 dr 3 2

(7.23) where σ is the surface energy of the cluster boundary that is equal to the free energy corresponding to the solution of the one-dimensional problem [24, 66, 78], i.e.,

  ∞  2 ∞  3 dϕ dr + Φ (ϕ) = dϕ 2Φ (ϕ). (7.24) σ= 2 dr 0 0 Here the integral should be calculated over the external field. The radius of the cluster can be obtained as Rc = 2σ/ε. In our case μ3 μ2 l ε = λ2με 1/2 and σ = 3λ , then Rc = 3λ1/2 ε and the effective value of the free energy variation from the cluster formation is given by ΔF =

8πσR2 . 3

(7.25)

The probability of the formation of one cluster may be written in the form     8πσR2 ΔF = exp − . (7.26) P (R) = exp − kT 3 The characteristic dimensions of the spatially inhomogeneous distribution of particle concentration in the end provide a criterion for the first-order phase transition with cluster formation. The criterion of instability given by this relation may be interpreted as a condition for the formation of a spatially nonuniform distribution for given temperature, which depends on the concentration of particles and on the characteristic length of the new structure. The void formation may be described in a similar manner by introducing new variables g(r) ≡ 1 − f (r) and φ = 1 − ϕ that describe the states of the system from the nonoccurence of particles at different points of the space. In terms of the new variable φ,

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

page 229

229

we have an expression for free energy similar to the previous form but with new coefficients, i.e.,   kT 2 , V = 4 U (ρ)dρ, μ  = V − c(1 − c)   = λ, and ε = ε−14 U (ρ)dρ which are determined by the previous λ coefficients. This renormalization is done to bring an additional part of the free energy ΔFφ rather than to change the nonlinear coefficients. In terms of the new function φ, we can obtain the size of a cell as a cluster of voids. In the present approach, the sizes of voids as a new phase of voids without particles may be obtained similarly c = 2 σ / ε. In our case to the previous case, we thus have R ε =

2 με , 1/4 λ

σ =

μ 3 , 3λ

then  22 c = μ l . R 1/2 3λ ε This formula suggests that the size of a void is much greater than the size of a cluster that can be formed in this system too. The probability of the formation of a void may be written in a similar form as with new variables from the thermodynamic description of the cellular structure formation. The size of a cell depends on the concentration of particles. This general formula for the probability of formation of voids without particles is similar to the formula that presents the general description of the cluster formation of the new phase. To conclude this section, we have independently estimated the spontaneous formation of loosely bound ordered aggregates of colloidal particles and possible formation of cellular structures due to different nature of interaction. Thus, phase separation in colloidal fluids is directly related to the percolation transition of the wetting solvent phase. The solvent induced phase separation is driven energetically. In the present approach, we can estimate the sizes of voids as the characteristic length of the instability of the nonuniform distribution of particles.

February 5, 2021

230

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

Liquid crystal colloids Now we employ the results of [77–79, 82–85, 89, 97–133, 146, 147, 154–159, 161–177, 179–193] to briefly present the picture of the cellular structure formation in liquid crystal colloids. Dispersed liquid crystals with macroscopic particles of a foreign substance are a particular feature of such systems. The particles form various structures as observed experimentally [183–185]. It is well-known that the order greatly influences the electro-optical and rheological properties of colloidal dispersions. The system of particles in a liquid crystal changes the liquid crystal state to the soft solid [180, 181] (Fig. 7.3). The liquid crystal exists in two phases — the isotropic phase, when we observe nonorientation ordering in the long axis of molecules, and the ordered phase, when orientation ordering occurs in the long axis of molecules of the liquid crystal. The particles introduced in such an isotropic phase of a liquid crystal can cause orientation ordering at the expense of the formation of a solvent area. The existence of such deformed areas leads to effective interaction, when these areas overlap [192, 195–197]. Such interactions exist at short distances. In the nematic phase, orientation ordering occurs at long distances. The deformation of the elastic field causes the longrange interaction between particles that are included in the nematic liquid crystal [187–194, 198].

Fig. 7.3. Cellular structure in system: hard spherical particle Zr02 in liquid crystal 8CB.

page 230

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 231

Cellular Structures in Condensed Matter

231

Actually, however, different interactions occur at different distances. At short distances, interactions are induced by the change of the scalar parameter while at long distances the interactions are induced by the deformation in the elastic director field. In the case of low concentrations, interaction between two spherical particles was obtained in [195–197]. For small nanometer-sized particles immersed in the liquid crystal phase, we may assume that each particle changes the order parameter. If the number of particles in the local area increases, then the change of the order parameter increases too. For this process, we cannot take into account the distributional change of the director field. For the case of high concentration of particles introduced into the matter, a more consecutive account of their collective effect is presented in [189, 192]. Interaction can cause new ordering in the system of introduced particles. The effect of ordering is important in the new phase formation in the substance. In this section, we consider the behavior of a small particle in a liquid crystal and obtain the size characteristic in the formation of a cluster as a function of the concentration and temperature of the medium. The distortions of the order parameter produced by a small particle can lead to the effective interaction between particles and thus motivate the segregation process. We show that this interaction is responsible for the “cellular” and cluster texture in a system of small particles immersed in a liquid crystal. In order to describe such a system, we start from the volume elastic free energy density in the form given by [135]  1 AQij (r)Qij (R) + LQij,k (r)Qij,k (r) , (7.27) fb = 2 where comma indicates both derivation and summation over the repeated indexes. Here A is a positive constant, because we describe the liquid crystal state after phase transition, and L > 0 quantifies the cost of creating a distortion in the nematic phase. For the sake of simplicity, we apply the one-constant approximation. Each particle locally changes the order parameter and this fact may be taken into account through the free energy density in the form 1 fc = Wij f (r)Qij (r), 2

(7.28)

February 5, 2021

232

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 232

Statistical Physics of Nonequilibrium Systems

where Wij in the general case is the tensor coefficient that determines the interaction between particles and the order parameter, and f (r) is the distribution function of particles in the liquid crystal. The order parameters in the volume may be written in the well-known form   1 (7.29) Qij (r) = S(r) ni nj − δij , 3 where n is the director, S(r) presents the order parameter in the volume of the liquid crystal. We can present the free energy for a many-particle system as given by    L 1 Fb = dr Qnj,k (r)Qij,k (r) + 2 Qij (r)Qij (r)+wij f (r)Qij (r) , 2 ξ (7.30) where the correlation length ξ 2 = L/|A| and wij = Wij /L is introduced to represent the normalized coupling constant. This free energy describes the behavior of the liquid crystal with particles. The changes caused by individual particles introduced in the medium are developed and produce the average field of deformations of the order parameter.This makes it possible to correctly take into account the collective effect of all particles and to find the self-consistent interaction in the case of high concentration. The minimum of the free energy δ δQij (r)

F =0

yields the Euler–Lagrange equation ΔQij (r) −

1 Qij (r) = wij f (r). ξ2

The solution of this equation can be presented in the form  Qij (r) = wij dr f (r )G(r, r ),

(7.31)

(7.32)

where G(r, r ) is the Green function of the previous equation. Its form is well known, i.e.,   |r − r | 1  exp − . (7.33) G(r, r ) = |r − r | ξ

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

page 233

233

This result presents the solution for the distribution of the order parameter by particles with the distribution function f (r) included in the liquid crystal phase. Now we substitute this expression for Qij (q) again in free energy and thus obtain the free energy produced by all particles in the liquid crystal as given by  2  Lwij (7.34) dr dr f (r)f (r )G(r, r ). Fb = 2ξ 2 This free energy is associated with the energy of interaction between areas with particle concentrations f (r) at different space points. The immersed particles change the order parameter as a function of concentration; the changes of the order parameter produce the effective interaction between them. Having found the elastic free energy of included particles, we can study the thermodynamic behavior of an aggregate of nanoparticles and describe the conditions for the formation of a new structure. The criterion of instability in the homogeneous distribution of particles can be interpreted as a condition for the formation of spatially nonuniform distribution for given temperature that depends on the concentration of particles and on the characteristic length of the new structure. In order to determine this condition, we have to add to the elastic free energy the entropy part that may be written in the standard as    f (r) ln f (r) + [1 − f (r)] ln[1 − f (r)]dr . (7.35) Fs = kT The entropy part of the free energy is responsible for the classical particles that do not occupy similar spatial positions. The minimum of the two-part free energy Fb + Fs is associated with the self-consistent field solution for the distribution function f (r). This function corresponds to the solution that describes some thermodynamic state of the particle arrangement phase. If their distribution is inhomogeneous, then the solution serves to find the stable phase associated with the nature of interaction and temperature. If the particle solution is disordered, then by definition, the mean value

February 5, 2021

234

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

f (r) = c, where c is the relative particle concentration. The concentration inhomogeneity gives rise to an additional term f (r) = c ±ϕ(r), where ϕ(r) is the change of the particle distribution function. If concentration inhomogeneities are smooth and their scale is much longer than the interparticle distance, then this quantity may be interpreted as the change of particle composition. We can write the free energy growth rate associated with the inhomogeneous particle distribution by the power series expansion and using the long-wavelength expansion of the concentration, i.e., 1 ϕ(r ) = ϕ(r) + ρi ∂i ϕ(r) + ρi ρj ∂j ∂i ϕ(r). 2 Here ρ = r − r is the distance between two different space positions. In this case, we may rewrite the free energy, that depends on the change of the distribution function of particles, in the standard form given by    1 dr l2 (∇ϕ)2 + μ2 ϕ2 + , (7.36) ΔF (ϕ) = 2 where μ2 ≡



kT +V c(1 − c)

,

V =

2 Lwij 2ξ 2

 ρG(ρ)dρ,

and 2 Lwij l = 2ξ 2 2



ρ3 G(ρ)dρ.

In our case, we can obtain μ2 ≈

kT 2 + 2πLwij c(1 − c)

and 2 2 ξ . l2 ≈ +12πLwij

This system is always unstable and the length of the first instability is  (7.37) λ = l2 /μ2 ∼ 2ξ

page 234

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

page 235

235

for kT 2 < 2πLwij , c(1 − c) ξ is the correlation length. Under the reverse condition, we observe the smallest length of the inhomogeneous particle distribution. In such a case, the Coulomb-like attraction between particles separated by the distance of few nanometers results in the formation of a cluster with strongly interacting particles [189, 192, 195–197]. In the liquid crystal phase, the correlation length is ξ ∼ 1μm and it is the average size of a cluster of segregated nanoparticles. Cluster formation in a system of nanoparticles is the natural result of the elastic interaction of areas with different particle concentrations. These particles produce the deformation of the elastic matter and this deformation leads to the segregation of small particles. This behavior motivates the long-range repulsive interaction and the short-range attractive interaction through the changes of the order parameter of the liquid crystal. The equilibrium distribution of particles corresponds to their spatially nonuniform distribution. This spatially nonuniform distribution of the introduced particles leads to areas free of particles. The mutual effect of particles and the medium results in the nonuniform distribution of cooperating particles. Thus some kind of a new soft body is formed, whose properties differ from the properties of the medium [180, 181]. Liquid crystal colloids in nematic phase In recent years, much attention has been paid to the study of the interaction and phase behavior of colloidal particles dispersed in the ordered-phase liquid crystal [78, 193–195, 198–202]. The cellular structure is observed only in the liquid crystal phase while in the isotropic phase, it is not observed. This visualization is based on the macroscopic sizes of immersed particles. To describe the peculiarities of the behavior of a many-particle system in a liquid crystal one should take into account the interaction via the director elastic field. As was shown previously [154, 198], a foreign particle causes the

February 5, 2021

236

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

liquid crystal distortion in a region much greater than particle dimensions and thus leads to an effective interaction with other similar particles via the director field deformation. This particle may also be regarded as a particle surrounded by a “solvating shell” provided the interaction between the particle and the liquid crystal molecules is much stronger than the intermolecular interaction responsible for the liquid crystal formation. The solvating formation may be regarded as a particle of size equal to the size of the deformation coating around the particle. Thus, its interaction with another similar formation may be described by the director field deformation [188]. The validity criterion of this treatment of spherical particles may impose the boundary condition on the director for such a formation. In this sense, the interaction of spherical particles is also related to the elastic director field deformation in the liquid crystal in which these particles are dissolved. The problem of interaction between the particles of the nematic liquid crystals has a solution in the self-consistent molecular field approximation when the director field distribution on the surface of an individual particle is determined by the joint effect of all other particles [191]. The effectiveness of such interaction is determined, first of all, by the geometric parameters of the inclusion, by the force of adhesion of the liquid crystal molecules to the surface of such an inclusion, and by the elastic properties of the matter. In the general case, the value and nature of the interaction are determined by the value and nature of the violation of the director distribution symmetry [191]. In the case of small spherical particles, the interaction energy is shown in many papers [25, 154, 198] to have the quadruple form, i.e., 2  3 − 35 cos 0θ + 35 cos4 θ 4π 4 W R0 . (7.38) W (ρ) = 15 ρ5 Here R0 is the particle size, ρ is the distance between particles, and θ is the angle between the director and the radius vector of the distance. As is shown in paper [154], the system of particles with interaction of this type is not stable. In what follows, we present some way of solving the problem of void formation in a system of colloidal particles in the liquid crystal, phase. In view of the anisotropic specifics of liquid

page 236

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 237

Cellular Structures in Condensed Matter

237

crystals, we can rewrite the change of the free energy in the form given by     2 1 dr ξ ∇ ϕ + ξ⊥ (∇⊥ ϕ)2 + μ2 ϕ2 , (7.39) F (ϕ) = 2 where ξ =

1 2



U (ρ)ρ 2 dr,

and ξ⊥ =

1 2



U (ρ)ρ⊥ 2 dρ,

allows for different probable lengths of the changing distribution of particles along and perpendicular to the director, i.e., ρ , and ρ⊥ are the components of the interparticle distance vector. The minimum of the free energy realizes a spatially inhomogeneous particle distribution, provided the sign satisfies some relation and the values of coefficients are determined by the interparticle interaction potential. In order to reveal the condition under which the homogeneous particle distribution becomes unstable, we should calculate all the coefficients [154]. Here, we consider the case when the concentration may be described by the uniform configuration. This concentration wave is formed in accordance with the new equation. We can write the concentration wave in the form of the Fourier transform  ϕ(k) = drϕ(r) exp(ikr) of an arbitrary function ϕ(r) defined in a finite volume with periodic boundary conditions. In this case, the relation that determines the inhomogeneity of the particle distribution is given by kT 2 = −μ2 − ξ k2 − ξ⊥ k⊥ . c(1 − c)

(7.40)

Thus, the homogeneous distribution of particles dispersed in a nematic liquid crystal can be unstable in a limited temperature region where the isotropic phase exists, and a spatially modulated distribution can be generated with the wave vectors k and k⊥ . In the liquid crystal phase, the estimation of the interaction energy yields the value V ∼ 104 kT that completely realizes the condition for the formation of a spatially inhomogeneous distribution of particles in this matter [193] (Fig. 7.4). Below we see that the

February 5, 2021

238

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 238

Statistical Physics of Nonequilibrium Systems

Fig. 7.4. The result of computer simulation of possible colloidal particle distributions confined to a cubic cell of nematic liquid crystal.

formation of voids in an isotropic phase is impossible while formation of voids in a liquid crystal phase is realized in many cases. The energy of interaction between particles in a liquid crystal is much greater than the energy of interaction in an isotropic phase. Feronematic colloids In this section, we briefly present our previous results [188, 191] on another type of liquid crystal colloids, where a cylindrical particle with an intrinsic magnetic moment is immersed into a liquid crystal. This magnetic moment induces interaction of the external magnetic field and the system of particles and can change the orientation of both the particles and the liquid crystal. In principle, however, the rest of the particles without additional moments should effect the interaction potential. The paper [188] examined this case and has shown that deformations from all particles lead to the exponential screening of the pair interaction potential and this fact can explain the cellular structure. Here, we reproduce this explanation. The interaction potential for a cylindrical particle may be written as [188] Upp = −

1 p p + exp(−ξρ) A AQ  2πK l l l,l ρ

(7.41)

l = p are presented only by the first terms A where operators A l αlm (n0 · km ) because all others give higher powers in 1/ρ [191]. Here Q+ l,l = (r3 · κ1 )(r1 · κl ) + (r2 · κl )(rl · κl ),

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 239

Cellular Structures in Condensed Matter

239

where κl represents the orientation of the particle. For the cylinder, the tensor  αlm = 2 dsW (s)νl (s)νm (s) has the components [191] α11 = α22 = dLπW, α33 = d2 πW , the others αlm = 0. Here L is the length of the grain. So α33 /α11 = d/L ∼ 0.4 and we should neglect α33 . Thus, the interaction energy may be rewritten as Ucyl (R) = −

α211 sin2 θ cos2 θ exp[−ξ(θ)ρ] . 4πK ρ

(7.42)

In the case of equilibrium orientations θ = 0, π/2 [203, 204], the screened Coulomb-like interaction does not occur and the higherorder terms remain. It is clear that the macroscopic particle concentration c causes screening  of the pair interaction potential with −1 the screening length ξ ≈ K/W cS (we imply that W is here the absolute value independent of the sign), S is the area of the particle. The screening occurs both for homeotropic and planar anchoring. The concentration here is included in the inverse screening length ξ only, so that in the limiting case c → 0, we have ξ = 0 that brings us back to the unscreened result. In order to describe the experimental results for the dependence of the field-induced birefringence on the strength of the applied field, concentration of the magnetic dopant, and the thickness of the nematic cell paper [188], we introduce the following free-energy density functional given by   1 kT f (r) ln f (R)dV + 2 f (r)f (r + ρ)U (ρ)drdρ. (7.43) F = v 2v Here we have to find a condition of stability loss in such a system of attracting particles. It means that the concentration reduces to f (r) = c + δf (r), where c is the ground concentration as before. We perform an expansion in series 1 f (r + ρ) ≈ f (r) + (ρ∇)f (r) + (ρ∇)2 f (r). 2

February 5, 2021

240

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 240

Statistical Physics of Nonequilibrium Systems

Thus we obtain 1 F − F0 = 2



N δf 2 (r) + M (∇δf )2 ,

where 1 N = 2kT /v + 2 v and 1 M =− 2 2v



∞



(7.44)

∞

U (ρ)dρ

(7.45)

R0

U (ρ)ρ2 dρ.

(7.46)

r0

Here r0 is as before the size of the particle. Given that U < 0, the phase transition occurs when N < 0. In this case, we have ξ R0 1 and hence we can obtain 4πe 12πe 2kT − 2 2 and M ≈ 4 2 . N≈ f0 v ξ v ξ v Below the critical point N∼

4πe . ξ 2v2

The length of the first instability is  1 linst = 2M/N ∼ . ξ Here ξ −1 (θ) =



(7.47)

K/c |a(θ)|

is the screening length. In the experiment [84] ξ −1 ∼ 60 μm as is found above. For the concentration c ∼ 1010 cm−3 , the average distance between particles is l ∼ 5 μm, so that ξ −1 l. There are about 1000 particles in volume ξ −3 and they induce screening indeed. As we have discussed before, linst ∼ 50 μm that is in good agreement with the experimental size of the “cells” [84]. In the experiments by Chen and Amler [84] with the cylindrical particles, the concentration is c ≈ 1010 cm−3 , S ≈ RL. The radius of the grain is R ≈ 0.05 μm, the length L ≈ 0.5 μm. The elastic constant K ∼ 10−7 dyn, the anchoring

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

page 241

241

energy W ∼ 10−3 dyn/cm, and we can find λres ≈ 60 μm. Both concentration and anchoring can be changed so that the resonance range is λres ∼ 1 − 100 μm. The cell formation in a system of colloidal particles in a liquid crystal can be responsible for the Janossy effect that consists of the decrease by two orders of the critical field of the Fredericks transition under laser illumination in a liquid crystal with J-aggregates as compared to a pure liquid crystal [205]. J-aggregates in a liquid crystal are oriented with respect to the minimum free energy. This orientation can be neither parallel or perpendicular to the director, being just an equilibrium orientation of individual aggregates. Under laser illumination, each aggregate changes its intrinsic orientation and can cause deformation of the director elastic field that gives rise to the interaction between aggregates as we have described above. This energy of interaction is shown for the formation of cellular structures in a system of aggregates. The formation of a cellular structure in a system of aggregates can be caused by the change of the orientation of the director that may be regarded as the Fredericks transition. 7.4 Geometry of the Distribution of Interacting Particles Any physical theory is based on the postulated geometric properties of space. The problem of geometry as a whole is equivalent to the problem of the behavior of the fields that form the space [206–210]. We think that in the problem formulated in such a way, the geometric aspect is important not only for the description of the Universe but also for the study of physical phenomena. In the last years, interest has been growing in the application of the methods of differential geometry in statistical physics and thermodynamics. The geometric approach is represented by two main ways of investigation. The first approach deals with the metric while the other one considers the contact structure of the thermodynamic phase space. Using the metric tensor we can calculate the scalar curvature for some statistical and thermodynamical models and

February 5, 2021

242

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

study the effects due to the relation between metric and physical quantities. Another idea is that the curvature is proportional to the reciprocal free energy of the system. The scalar curvature is divergent near the critical point and, due to its relation to the second moments of fluctuations, the stability of the systems can be measured [211, 212]. The average degree of instability of Hamiltonian dynamics can be given by the curvature-related quantities integrated over the whole mechanical (Riemannian) manifold. This fact establishes a link between the dynamical aspect of a given system, the stability or instability of its trajectories, and some global geometric properties of its associated mechanical manifold [213]. Temperature dependence of abstract geometric observables, e.g., averages of curvature fluctuations, has been studied [214] by the Riemann geometrization of Hamiltonian dynamics, where Lyapunov exponents are related to the average curvature properties of submanifolds of the configuration space [215]. Deformations of submanifolds of thermodynamic equilibrium states introduced by continuous contact maps on a phase-space manifold have been considered via the geometric formulation of thermodynamics [216]. In Riemannian geometric approach to thermodynamics, the theory of fluctuations is also considered [217]. In this approach, for the case of two independent thermodynamic variables, the Riemann curvature is assumed to be inversely proportional to the free energy near the critical point [218]. That leads to a partial differential geometric equation for free energy. Following [218], the solution is a generalized homogeneous function of its arguments and specifying the values of critical exponents results in a full-scale equation of state. Later, this postulate was generalized for the cases with more than two independent thermodynamic variables [219]. In condensed matter physics, attempts have already been made to use the geometric approach for the description of the phase space [211–213, 215, 220–222], to find the criteria of phase transitions [78, 211, 212, 221], to study the formation of the phase boundary [223] and the membrane [224]. The geometric aspect of the physical phenomena occurring in condensed matter may be important for

page 242

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

page 243

243

the study of structural phase transitions accompanied by the formation of nonuniform distributions of interacting particles [3, 59]. A nonuniform particle distribution can induce an individual effective space that reflects the character and intensity of the interaction in the system. In the Euclidean space we observe particle distributions determined either by the structure of the crystal in which redistribution of spins, dipoles, atoms, molecules occurs, or by the geometry of the sample with fixed boundaries in which the macroscopic structures are formed. But the effective geometry of the particle distribution can be other than Euclidean. This observation is especially important for soft systems with particle distributions governed only by the nature and intensity of the intrinsic interaction. Even in the crystal structures formed by the interaction between atoms, an effective nonEuclidean geometry can occur and be manifest in the existence of quasicrystals, fullerenes etc [225]. Some approaches have been proposed to use the geometric aspects for the description of physical problems of condensed matter. First of all, it is the usage of the direct analogue with the general theory of relativity. Another approach is based on the calculation of the distribution of interacting particles in the curved space. In biophysics and the theory of liquid crystals, we have solved the problem by obtaining new formations from the minimum of the free energy. Nevertheless, very important questions remain unsolved concerning probable intrinsic reasons that induce the changes of the internal geometry in the system of interacting particles as well as how the nature and magnitude of the interaction between particles change this geometry. The pure gravitational interaction in the system of self-gravitating particles leads to the fractal distribution of particles in the space [8]. Many physical and chemical systems (e.g., membranes, vesicles, type I superconductor films) displaying macroscopic patterns and textures in equilibrium have been analyzed within the framework of competing interactions. The related phenomena of structure formation in coloidal suspensions and superlattice formation in adsorbate films on crystalline substrates have also been studied [226]. In condensed matter physics, the phase transitions with the formation of

February 5, 2021

244

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 244

Statistical Physics of Nonequilibrium Systems

spatially inhomogeneous distributions of particles, clusters or cellular structures are of great interest. It was shown earlier that structure formation in all systems of interacting particles is of similar physical nature [227] and hence we may describe them geometrically in the same way. The purpose of this section is to find the effective space of a thermodynamically stable distribution of interacting particles and to give a geometric description of structural phase transitions induced by the nature and intensity of the interaction in the system. Thus we consider a system of many particles. The free energy in the self-consistent field may be written in the many-body approximation as F = Fp + Fs + Fn . Here

 Fp =

U (r − r )f (r)f (r )drdr + · · ·

(7.48)

(7.49)

is the free energy in terms of the particle distribution function f (r), where U (r − r ) is the sum of all energies of particle interactions,  (7.50) Fs = kT (f (r) ln f (r) + [1 − f (r)] ln[1 − f (r)]) dr is the entropy part of the free energy,   W (ri − r)dr Fn = f (r)

(7.51)

i

is the free energy resulting from coupling between particles and matter where W (r) gives the microscopic information about wetting properties of the surface of a particle located at the space point ri . The minimum of the free energy (7.48) corresponds to the selfconsistent field solution for f (r). If the distribution of particles in the system is homogeneous at each space point, ri , then f (ri ) = c where c = const is the particle concentration. In the case of inhomogeneous particle distribution f (r) = c + ϕ(r), where ϕ (r) is the deviation of particle concentration from equilibrium at different space points. In the continuum description (when the deviations of c are smooth on

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 245

Cellular Structures in Condensed Matter

245

the scale much longer than the distance between particles) we can write ϕ (r) as power series expansion 1 ϕ(r ) = ϕ (r) + ρi ∂i ϕ(r) + ρi ρj ∂j ∂i ϕ(r) + · · · , 2

(7.52)

where ρ = r − r is the distance between two particles. In the longwavelength approximation, using the expansion (7.52) and taking  into account that f (r)dr = N , ϕ(r)dr = 0, we may rewrite the free energy (7.48) as    1 2 2 1 4 1 2 2 l (∇ϕ) − μ ϕ + λϕ − εϕ , (7.53) ΔF (ϕ) = dr 2 2 4 where the coefficients kT , μ =V − c(1 − c) 2

 V =

U (ρ)dρ,

2



l =

U (ρ)ρ2 dρ

(7.54)

are determined in terms of the energy of interaction between the particles in the system. Coefficient λ is responsible for the nonlinearity of the system induced by the many-body interaction. For example, λ ∝  U (ρ)U (ρ )dρdρ , where ρ is the distance between two individual particles and ρ is the distance between other individual particles. The nonlinearity may be employed to determine the geometry of the effective spatial distribution of interacting particles. The last coefficient ε = N 4πR02 W represents the energy that includes each particle of size R0 through the wetting effect. The most important contribution to the concentration is associated with the field configuration for which the value of the free energy (7.53) is minimum, i.e., Δϕ −

dΦ = 0, dϕ

(7.55)

where the potential Φ is given by 1 1 Φ = − μ2 ϕ2 + λϕ4 − εϕ. 2 4

(7.56)

February 5, 2021

246

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

When the difference of the effective potential minimum values is greater than the barrier height, then the free energy may be written as



 ∞ 1 dϕ 2 4π 2 r dr + Φ (ϕ) = − r 3 ε + 4πr 2 σ, (7.57) ΔF = 2π 2 dr 3 0 where σ is the surface energy of the cluster boundary that is equal to the free energy associated with the solution of the one-dimensional problem [65, 78], i.e.,

  ∞  2 ∞  3 dϕ dr + Φ (ϕ) = dϕ 2Φ (ϕ). (7.58) σ= 2 dr 0 0 Each particle distribution formation in condensed matter induces an individual effective space that does not occur in the Euclidean case. The effective space can be observed through the curvature, torsion, and probable realization of the topologies with arbitraryorder symmetries (including fifth, seventh, etc.). The symmetry and topological properties of the effective space are determined first of all by the specifics of the interaction in the system. It is clear that the problem becomes much more complicated, nevertheless the way to solve it still exists. The idea is associated with the method of constrained global optimization for the Thomson problem of charged particles arrangement over a sphere [228–230]. Having solved this problem, one finds the configuration of the deformed lattice drawn on the surface of a given sphere. Here it is appropriate to mention the system of electrons on the surface of liquid helium. The structure of negative ions in liquid helium after the insertion of electrons in helium was original and unusual. Such particles, being localized inside a spherical void, create a bubble in the liquid helium [231]. The solution of the problem of how electrons are located over the sphere is equivalent to the solution of Thomson problem when N point charges on the surface of a unit conducting sphere interact only through their mutual Coulomb forces and try to find the configuration of charges for which the Coulomb energy is minimized. The system “helium + charges” is demonstrative in the study of nonlinear effects. Electrons repulse each other hence we have a

page 246

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

page 247

247

curved space. The many-body interaction in a system of particles determines the nonlinearity in the free energy. The geometry of the space, in turn, could be determined by the interaction. Thus, the nonlinearity in free energy leads to the inhomogeneous geometry of the space that is non-Euclidean. The minimum of the free energy corresponds to a spatially inhomogeneous particle distribution only for specific signs of the coefficients. Both signs and values of these coefficients are determined by the interparticle interaction [227]. The symmetry order is varied including the fifth-order case. This configuration corresponds to the energy minimum. Obtaining the lattice configuration of particle arrangement over the spherical surface that determines the global energy minimum is in principle similar to the case of particle distribution in the Euclidean space [228, 230]. Different lattice configuration on the spherical surface makes it possible to realize the symmetry order that cannot be observed in the usual Euclidean space [225]. This provides a possibility to realize an arbitrary geometry of interacting particles arrangement determined by the interaction character and intensity without additional restrictions of external conditions. A requirement arises to describe a new geometric background that is determined by the distribution of interacting particles and replaces the Euclidean space. This effective geometry should realize the free-energy minimum with regard for the curvature, torsion, or other geometric characteristics induced by the interaction in the system. Demonstrative examples of twisted spaces in the condensed matter are the helical structure of magnetic substances, cholesteric liquid crystals, and DNA, where the character of the microscopic interaction gives rise to a new topological invariant that in the macroscopic representation induces the effective twisted space with the helix proportional to the interaction parameter. The system of electrons on the surface of helium may be a physical example of such type of changes of the geometry. Each electron deforms the surface because electrons interact not only by the Coulomb repulsion but also by the elastic field of deformations. This interaction is attractive and together with the Coulomb repulsion causes the formation of a spatially inhomogeneous distribution of particles that forms a new surface. It is an effective geometric

February 5, 2021

248

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 248

Statistical Physics of Nonequilibrium Systems

background. As it is known, in the case of high concentration, electrons form bubbles of a new phase [111]. Such formation may be described in geometric terms. So, the main problem is to find the geometric characteristics of the effective space in terms of the physical parameters of the interaction nature and intensity and to apply them to describe various geometric structures [232]. Geometric characteristics of the space are determined by the minimum of the free energy of the system. The presence of geometric characteristics visualizes probable structures occurring in the particle distributions and, moreover, makes it possible to give an adequate description of physical effects associated with the interconversion thereof. The most important contribution to the concentration is associated with the field configuration for which the value of the free energy (7.53) is minimum, i.e., Δϕ −

dΦ = 0, dϕ

(7.59)

where the potential Φ is given by 1 1 Φ = − μ2 ϕ2 + λϕ4 − εϕ. 2 4

(7.60)

When the difference of the minimum effective potential values is greater than the barrier height, then the free energy may be written as



 ∞ 1 dϕ 2 4π 2 r dr + Φ (ϕ) = − r 3 ε + 4πr 2 σ, (7.61) ΔF = 2π 2 dr 3 0 where σ is the surface energy of the cluster boundary that is equal to the free energy associated with the solution of the one-dimensional problem [65, 78], i.e,

  ∞  2 ∞  3 dϕ dr + Φ(ϕ) = dϕ 2Φ(ϕ). (7.62) σ= 2 dr 0 0 If the geometry of particle arrangement is changed because of substance redistribution, which is equivalent, say, to the deformation of the initial lattice, then it is necessary to consider the value of this

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

page 249

249

deformation. By analogy to the theory of gravity, we have to allow for the space deformation in the description of the free energy. In most cases the potential energy U (r − r ) depends on the relative particle position r − r . It is more interesting, however, to consider the dependence of the potential energy on the real positions rather that an arbitrary coordinate system. The uncompensated interaction energy associated with the relative particle arrangement can produce the deformation of the initial lattice that provides the free-energy minimum for these deformations. The interaction between particles is assumed to be so strong that their rearrangement produces lattice deformations compensating for the excessive disproportion. Then the requirement arises that these deformations should be in correspondence with the free-energy minimum calculated for these deformations. In this case, all factors of the phenomenological representation of the free energy should depend on the space point where they change the macroscopic characteristics of the system. Even in the case of linear initial potential there is a nonlinearity in the potential part of the free energy and naturally there is an interaction of the geometric characteristics with the order parameter that can play the role of concentration and it is possible to describe the phase transitions in condensed matter of various types. Most importantly, is the dependence of a space point on the value of the order parameter. The description is more pictorial for a continuously distributed substance because in this case one can find the analogies with and differences from the gravity theory. For metric spaces, the procedure involves the metric tensor. The interparticle interaction in the system results in the redistribution of particles that reflects the geometric nature of the space. In this sense, we see the complete analogy to the gravity theory where the distribution of the fundamental scalar field, which in our case, has the meaning of the particle distribution function, forms the very space. The space curvature can be considered in terms of the free energy in the curvilinear coordinate system, i.e.,    √ 1 μν −gdV R + γg ∇μ ϕ∇ν ϕ + V (ϕ) , (7.63) F (ϕ, gμν ) = 2

February 5, 2021

250

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 250

Statistical Physics of Nonequilibrium Systems

where the metric tensor gμν determines the distance between particles that depends on the behavior of the interaction in the system. The free energy density of the background of the uniform distribution V (ϕ) serves for the “cosmological” constant. The space elasticity γ is completely determined by the interaction in the system. Here, we do not specify the form of the potential V (ϕ). But all coefficients contained in it, determined by the energy of interaction between particles, should depend on the space point and the order parameter in this spatial position [3, 59, 68]. The free energy (7.63) written in the curvilinear coordinate system provides a better description of the real physical situation. The curvature tensor is completely determined by the particle distribution [233]. And from the Einstein equation 1 Rμν = Tμν − gμν T 2 we get the relation Rμν = γ∇μ ϕ∇ν ϕ − gμν V (ϕ),

(7.64)

where Tμν is the energy-momentum tensor of the scalar field ϕ. The correlation of the particle distribution in the system with the space in which it occurs may be obtained from the free-energy minimum with respect to the particle distribution, i.e.,   dV (ϕ) 1 ∂ μν √ ∂ϕ = 0. (7.65) g g − √ g ∂xμ ∂xν dϕ The first integral of this equation does not present the free energy in complete analogy with the free energy in the general theory of relativity where we have  √ −g dV {R + 2V (ϕ)}. (7.66) F (ϕ) = Now the equations for the particle distribution geometry can be obtained by minimizing the free energy (7.66) with respect to

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 251

Cellular Structures in Condensed Matter

251

the metric, i.e., δF (ϕ) = δgμν



1 μν g R − Rμν 2

+

δV = 0. δgμν

(7.67)

In this representation, the scalar curvature R is proportional to dV (ϕ) = μ2 + λϕ2 , dϕ2

(7.68)

that completely confirms the results in [211, 213] and [214]. If the particle distribution is found, we can investigate the geometry of this distribution. The picture is even more clear physically in the case of a conformal space when the metric tensor is found by multiplying the Euclidean space metric by a scalar function. In the Euclidean space the distribution of particles produces the effective space that corresponds to the conformal metric tensor. Suppose that the suitable conformal transformation is given by gμν = ϕ2 hμν , where hμν is the metric tensor of the Euclidean space. Then the curvature tensor is described by the expression R0 = −6ϕ−3 Δϕ, where Δ is the Laplace operator. The free energy for the conformal space may be written as    2  √ √ ϕ R0 + V (ϕ) −g {R + 2V (ϕ)} dV ≡ γ gdV, F (ϕ) = 12 (7.69) where ˜ μν = 1 hμν R, R 48

2  = gμν Rμν = ϕ R0 R 12

for the observation that ϕ2 = hμν gμν and gμν gμν = 4 here. Multiplying by gμν yields 1 dV (ϕ) ϕ2 R + = 0, 6 ϕ dϕ which is similar to γΔϕ −

dV (ϕ) =0 dϕ

(7.70)

February 5, 2021

252

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

(see also (7.55)) and finally reduces to the particle distribution function in the usual Euclidean space. When we substitute the solution of Eq. (7.70) in the expression of free energy, we can determine the condition for the formation of a new phase. In the beginning, the distribution of particles is homogeneous in the Euclidean space while the bubble thus formed is a nonEuclidean object. The flat surface where the particles are placed becomes spherical with even different concentration of particles. The formation of a bubble may be presented as a change of space topology, at the expense of the change of curvature with different particle concentrations on concentric surfaces around the center. If we know the dependence of the radius of a separate concentric sphere on particle concentration, we find the dependence of curvature on concentration. In other words, the distribution of particles in a radial direction can be interpreted as the dependence of the curvature radius on concentration. The solution of the field equation determines the effective space that corresponds to the particle distribution. Vesicles are often used as model cell membranes. When we study the forms of membranes we consider the simplified free energy in the two-dimensional case, i.e.,    √ 1 μν −gdS R + γg ∇μ ϕ∇ν ϕ + V (ϕ) . (7.71) F (ϕ, gμν ) = 2 The magnitude of the field is a function only of the polar angle θ. Because of the symmetry, the order parameter ϕ(θ) and the shape function R(θ) can be expanded in the associated Legendre series [234]. We thus have ϕ(θ) =

∞ 

1 ϕ2j−1 P2j−1 (cos θ),

j=1

⎛

R(θ) = R0 ⎝1 +

∞ 

⎞ r2j P2j (cos θ)⎠ .

j=1

In the case of equilibrium, the free energy is a minimum with respect to R0 , ϕ2j−1 , r2j . The contributions to the free energy from the

page 252

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

page 253

253

curvature and gradients in ϕ are independent in the region A = √ −gdS. The remaining part of the free energy applied in [234] may be written as     √ 1 1 2 2 1 4 μ ϕ + λϕ −gdS . (7.72) Fϕ = A  √ −gdS 2 4 It determines the change of the vesicle shape near the transition. In the disordered phase, the vesicle is assumed to have a spherical shape. Near the transition j = 1 modes dominate. The coupling of the shape to the order in the membrane leads to a continuous change in the shape and the sphere turns into cylinder, thus changing the topology of the space. For a given shape R(θ), minimization of the free energy with respect to ϕ(θ) is equivalent to solving a nonlinear differential equation for ϕ(θ). Its solution may be closely approximated by the parameter variation function as in the case of a bubble. In a simple model for tangent-plane order in the vesicle with spherical topology, it was shown by the mean-field theory how the development of such order leads to the change in the shape of the vesicle [234]. The coupling between the in-plane order and the Gaussian curvature results in the continuous change of the vesicle shape (from spherical to cylindrical) with the increase of the in-plane order degree. The illustration of the effective curved three-dimensional space could be realized by almost crystal structures called quasicrystals. One can form a perfect icosahedral crystal in the curved threedimensional surface of S 3 , a sphere in four dimensions. The frustration arises when we flatten the sphere to fill the flat space [225]. The frustration of the double-twist vector field in R3 is relieved on S 3 and thus gives rise to polytope-like models of the blue phases of cholesteric liquid crystals. The twisted vector field on S 3 also serves as a model for molten polymers. Polytopes may be thought of as crystals in the non-Euclidean three-dimensional space. Alternatively, we may consider them to be embedded in a higher-dimensional flat space. Lower-dimensional analogues of polytopes are polyhedrons and polygons that may be regarded as, respectively, two- and onedimensional non-Euclidean crystals embedded in three-dimensional

February 5, 2021

254

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

and two-dimensional flat spaces. Since the non-Euclidean space S 3 represents the surface of a sphere in the Euclidean four-dimensional space, we may do analytic geometry in S 3 using four-dimensional Cartesian coordinates. Following hereinafter [225], we admit that any point in R4 may be expressed as u = ua ea where ea are four basis vectors in R4 . We can easily calculate the geodesic separation between any two points on S 3 . Formally, it is possible to generate a three-dimensional rotation using quaternions that are described by the unit vector in S 3 . We have the coordinate system in S 3 in which any point written in the quaternion form corresponds to the threedimensional rotation. It is useful to label the points in S 3 to describe the polytopes. Certain four-dimensional rotation may be expressed clearly in terms of the familiar three-dimensional concepts. Ginsburg–Landau model of frustrated icosahedral order provides a phenomenological description of supercooled liquids, glasses and solid phases that may be crystalline or quasicrystalline. To construct a free energy for a liquid we have to identify an order parameter that might be nonuniform in the space. To construct a translation and rotation invariant free energy we have to consider the transformation properties of the order parameter. As in [225], when we combine a spatial translation r → r + u and rotation r → r + θ × r, we get the order parameter ϕq (r) → exp(iq · u + iq · θ × r)ϕq (r). To define an order parameter of this type for the polytope-like ordering we have to find an analogue of the reciprocal lattice vector for the polytope. The projection onto hyperspherical harmonics on S 3 is analogous to the Fourier transformation in R3 . Thereby for the density ϕ(u) on S 3 we have  (7.73) Qn,m1 ,m2 = dΩu Yn,m1 ,m2 (u)ϕ(u). The index n may be identified with the wave-vector q through the relationship |q|2 = k2 n(n + 2) where k = π/5d is the curvature of S 3 and d is the average distance between atoms in the icosahedral domains. In the paper [235], one can find the study of the universal property of curvatures in surface models that display a flat phase and rough phase, the criticality of which is described by the Gaussian

page 254

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 255

Cellular Structures in Condensed Matter

255

model. The Hessian relation gives a clear understanding of the universal curvature jump at roughening transition and facet edges. It also provides an effective way of locating the phase boundaries. In the terms introduced in [235] the universal relation between the surface curvature of the equilibrium, crystal shape, and Gaussian coupling constant yields k = (πkT )/(4σd). The total free energy associated with ϕq (r) consists of the transition–invariant gradient term so that the transition to the true ground state is likely to be of the first order, and the potential terms that begin at the quadratic order in Qn . Then [225]  K  |(∇ − iq × θ)ϕ|2 + rϕ2 + O(ϕ4) ΔF = 2 q =

1 Kn |DQn |2 + rn Q2n + · · · , 2 n

(7.74)

where ΔF = F − F (c), DQn = ∇Qn,m1 ,m2 − ik

 j

such that

ej



(Lnj )m1 m2 m m Qn,m ,m

m1 ,m2

1

2

1

2

⎧ ⎫ ⎨  ⎬ Lnj uj Qn (r). Qn (r + u) = exp i ⎩ ⎭ j

The spikes in the structure function occur for wave numbers corresponding to the negative coefficient of the quadratic term. The ground state of the free energy (7.74) quite likely corresponds to the polytope-type crystals. The general fourth-order theory is considered that combines the translation and orientation order parameters. It is also shown that quasicrystalline and icosahedral liquid-crystal phases occupy a large portion of the phase diagram [225]. Now, we shall show that a non-Euclidean geometry of particle distributions can be realized even in the absence of nonlinearity, For illustration, we consider a conformal space. We can investigate

February 5, 2021

256

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

page 256

Statistical Physics of Nonequilibrium Systems

the distribution of the interacting particles in the Euclidean space solving the appropriate equation. We rewrite the free energy in the cylindrical coordinate system and assume that a spatially modulated structure is realized along the OZ axis of this coordinate system. We have to find how the particle distribution profile varies in the direction perpendicular to OZ. The unknown function ϕ describes the deformation of the surface where the particles are distributed. The equation for the distribution function is equivalent to the equation for the surface profile in the direction perpendicular to the selected axis. In our case, this equation is given by  1 ∂2ϕ 1  2 ∂ 2 ϕ 1 ∂ϕ + k − μ2 ϕ = 0, + − 2 2 2 ∂r r ∂r r ∂θ γ

(7.75)

where r is the radius-vector perpendicular to the axis, θ is the polar angle, k is the wave vector of the periodic modulation along the axis. The exact solution of Eq. (7.75) is given by & $% 2 − μ2 √ k cos mθ ϕ(r) = C1 rI(1/2)√1+m2 γ √ + C2 rK(1/2)√1+m2



%$k 2 − μ2 γ

& cos mθ,

(7.76)

where Iν , Kν are the Bessel functions of reduced arguments. It is clear from the solution that the curved surface can realize an arbitrary-order symmetry  m of particle arrangement and various values of the parameter (k 2 − μ2 )/γ. The depressions and humps correspond to the probabilities of particle presence or absence. All information about the surface profile deformation that reflects the effective particle arrangement is determined only by the nature and intensity of the interaction in the system by the coefficients μ, γ, etc. This is a probable way of effective space formation in a system of interacting particles. Physically, this fact is equivalent to the existence of effective forces responsible for the elasticity of the space realized by the given particle distribution Fig. 7.5. In this sense our result corresponds to the results of [213] associated with the geometric description of phase transitions.

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Cellular Structures in Condensed Matter

Fig. 7.5.

page 257

257

Distribution of particles on the curved surface.

Thus to conclude, the approach proposed in this section makes it possible to find the effective space by the particle distribution in the system and to describe the visual experimental results in geometric terms. It might be important for the description of quasicrystal states with symmetry of the order such that it cannot be realized in the usual Euclidean space. This requires just allowing for the deformation of the usual crystal lattice or of the surface over which particles with relevant interaction are distributed. Moreover, a possibility arises to describe in geometric terms the structural formation in the macroscopic scale and to investigate their interconversion accompanied by modifying the topology of effective spaces induced by the arrangement of interacting particles. An approach to the description of the physical characteristics of condensed matter in geometric terms is thus offered. Probably in this way, it might be possible to describe the formation of quasicrystals as usual crystals but in the curved three-dimensional space. Moreover, it is possible to describe in geometric terms the phase transition by the creation of a spatially nonuniform distribution of the order parameter. As illustration of the dependence of an effective space curvature on the concentration of

February 5, 2021

258

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch07

Statistical Physics of Nonequilibrium Systems

interacting particles can be found in the experimental paper [232]. We suppose that the similarity to the general theory of relativity may not only be formal but to have deep origin, as the elasticity of the effective space always may be written in terms of the interpartical interaction. Such effective space depends on the conditions of the system of interacting particles along with their concentration and other average values associated with the macroscopic order parameter that changes the internal geometry of the condensed matter.

page 258

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Chapter 8

Statistical Description of Nonequilibrium Systems

8.1 Nonequilibrium Gravitating Systems This part of the book is devoted to a new approach proposed earlier in [1]. It deals with the nonequilibrium statistical operator that is suitable for describing gravitating systems. The equation of state and all thermodynamic characteristics needed for the description are determined by the equations that govern the dominant contribution in the partition function. The approach describes the real inhomogeneous distribution of particles and determines the thermodynamic parameters in a self-gravitating system. The main idea of this section is to provide a detailed description of a self-gravitating system based on the principles of nonequilibrium statistical mechanics and to obtain probable distributions of particles and temperature for fixed number of particles and energy of the self-gravitating system. Phenomenological thermodynamics is based on the conservation laws for average values of the physical parameters, e.g., the number of particles and energy. Statistical thermodynamics of nonequilibrium systems is also based on the conservation laws, however, for specific dynamical variables rather than for their average values. It presents local conservation laws for the dynamical variables. In order to find the thermodynamic functions in the case of a nonequilibrium system we have to rely on relevant statistical ensembles making allowance for the nonequilibrium states of the systems, under consideration. To describe the nonequilibrium stationary states of the system,

259

page 259

February 5, 2021

260

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Physics of Nonequilibrium Systems

we apply the concept of the Gibbs ensemble. In this case, the nonequilibrium ensemble may be determined as a set of systems contained under similar stationary external actions. Such systems show similar character of contact with the thermostat and possess all probable values of thermodynamic parameters corresponding to the given conditions. In the systems that are under similar stationary external conditions, a local equilibrium stationary distribution is formed. If the external condition depends on time, then local equilibrium distributions are not stationary. Exact determination of a local equilibrium ensemble requires the determination of the distribution function or the statistical operator of the system [96]. Finally, we remind that stable states of the systems of equilibrium of classical self-gravitating particles are only metastable because they correspond to the local maxima of the thermodynamic potential. Under the assumption that nonequilibrium states of a system may be determined by the inhomogeneous distribution energy H(r) and the number of particles (density) n(r), the statistical operator of a local equilibrium distribution for a classical system may be written in the form [96]     (8.1) Ql = DΓ exp − (β(r)H(r) − η(r)n(r)) dr , where the microscopic particle density may be given by the standard  expression n(r) = i δ(r−ri ). The integration in this formula should be carried out over the whole phase space of the system. It should be mentioned that the Lagrange multipliers β(r) and η(r) in the case of local equilibrium distribution are functions of the space point. The local equilibrium distribution can be introduced only if the relaxation time in the whole system is greater than the relaxation time in any part of this system. Once the nonequilibrium statistical operator is determined, we can obtain all thermodynamic parameters of the nonequilibrium system. For this purpose, we determine the thermodynamic relation for the inhomogeneous systems. The variation of the statistical operator by the Lagrange multipliers yields the required thermodynamic

page 260

February 5, 2021

Applications of Field Theory Methods. . . - 9in x 6in

14:3

b4116-ch08

Statistical Description of Nonequilibrium Systems

page 261

261

relations of the form [96] −

δ ln Ql = H(r)l δβ(r)

δ ln Ql = n(r)l . δη(r)

and

These relations make a natural general extension of the well-known relation for the equilibrium systems in the case of inhomogeneous systems. Conservation of the number of particles and energy of the system  may be described in terms of natural relations n(r)dr = N and H(r)dr = E. Further statistical description of nonequilibrium systems requires the knowledge of the Hamiltonian of the system. In the general case, the Hamiltonian of a system of interacting particles is given by H=

 p2 1 i + W (ri rj ). 2m 2 i

(8.2)

i,j

Such presentation of Hamiltonian is valid for spaces of various dimensions. In the three-dimensional case of a real space, the gravitation interaction energy may be written in the well-known form (r, r ) =

Gm2 , |r − r |

where G is the gravitation constant and m is the particle mass. In what follows, we write the energy density of a self-gravitating system in the form given by 1 p2 (r) n(r) + H(r) = 2m 2



W (r, r )n(r)n(r )dr .

(8.3)

The nonequilibrium statistical operator for the self-gravitating system is given by    2  p − η n(r)dr β Ql = DΓ exp − 2m   1    W (r, r )n(r)n(r )drdr . (8.4) − 2

February 5, 2021

262

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Physics of Nonequilibrium Systems

Having integrated over the phase space we have 1

dri dpi DΓ = (2π)3 i and all the thermodynamic variables β(r), p(r), and η(r) are functions of the space point. In the next step, we employ the wellknown method of field theory. Recently [3, 8], it was shown that the description of this system is exactly equivalent to the field theory of a single scalar field ϕ(r) that contains the same information as the original distribution function, i.e. all information on probable spatial states of the system. In order to perform formal integration in the second part of this presentation, we introduce additional field variables within the context of the theory of Gaussian integrals [7, 12], i.e.,   2 ν    βω(r, r )n(r)n(r )drdr exp − 2  2  ν βω −1 (r, r )ϕ(r)ϕ(r )drdr = Dϕ exp − 2    −ν βϕ(r)n(r)dr , (8.5) where

Dϕ = 

s

dϕs

det 2πβω(r, r )

and ω −1 (r, r ) is the inverse operator that satisfies the condition ω −1 (r, r )ω(r , r ) = δ(r − r ). Thus, the interaction energy is represented by the Green function for this operator and ν 2 = ±1 depending on the sign of the interaction or the potential energy. The field variable ϕ(r) contains information similar to the original distribution function, i.e., complete information concerning probable spatial states of the system. After the present manipulation, the field variable ϕ(r) contains the same information as the original distribution function, i.e., all information about probable spatial states of the system. The inverse operator W −1 (r, r ) of the gravitation

page 262

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Description of Nonequilibrium Systems

page 263

263

interaction in the continuum limiting case should be treated in the operator sense, i.e., W −1 (r, r ) = −

1 Δr δ(r − r ), 4πGm2

(8.6)

where Δr is the Laplace operator in real space. After this manipulation, the statistical operator may be rewritten in the form   p2 (r) + η(r) −β(r) Ql = DΓDϕQϕ exp 2m    (8.7) + β(r)ϕ(r) n(r)dr , where

 Qϕ = exp −

1 8πm2 G



 (∇ϕ(r)) dr . 2

(8.8)

The above-mentioned functional integral can be integrated over the phase space. Using the definition of the density, we may rewrite the nonequilibrium statistical operator as     p2 Ql = DϕDΓξ(ri ) exp − β(ri ) i − β(ri )ϕ(ri ) Q(ϕ(ri )), 2m (8.9) where a new variable ξ(r) ≡ exp η(r) is introduced, that may be interpreted as the chemical activity, and the definition 

1 dri dpi DΓ = 3 (2π) N ! i is used. Now it is possible to perform integration over the momentum. The nonequilibrium statistical operator takes the form    1

2πm 3/2 dri ξ(ri ) Ql = DϕQ(ϕ(ri )) N! 3 β(ri ) i   × exp β(ri )ϕ(ri ) (8.10)

February 5, 2021

264

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Physics of Nonequilibrium Systems

that may be simplified to yield     2πm 3/2 (∇ϕ(r))2 + ξ(r) − Ql = Dϕ exp 8πm2 G 3 β(r)    × exp β(r)ϕ(r) dr . (8.11) In the case of constant temperature β and absolute chemical activity ξ the statistical operator completely reproduces the equilibrium canonical partition function [3, 8]. Following [3, 8], we reduce the nonequilibrium statistical operator to the form     (8.12) Ql = Dϕ exp −S ϕ(r), ξ(r), β(r) where the effective nonequilibrium “local entropy” is given by    2πm 3/2 1 2 (∇ϕ(r)) − ξ(r) S= 8πm2 G 2 β(r)    × exp β(r)ϕ(r) dr. (8.13) Here we have used the definition of the chemical activity in terms of the chemical potential ξ(r) ≡ exp η(r)) ≡ exp(μ(r)β(r)) that depends on the space point. We have already mentioned that in the case of constant temperature β and absolute chemical activity ξ, the statistical operator completely reproduces the equilibrium canonical partition function [3, 8]. The saddle point method can now be employed to find the asymptotic value of the statistical operator Ql for N tending to ∞; the dominant contribution is given by the states that satisfy the extreme condition for the functional. It is not difficult to see that the saddle point equation presents the thermodynamic relation and may be written in a modified form, i.e., as an equation for the field variable (δS)/(δϕ(r)) = 0, the normalization condition  δS δS =− ξ(r)dr = N, δ(η(r)) δ(ξ(r))

page 264

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 265

Statistical Description of Nonequilibrium Systems

265

and the conservation law for the energy of the system  δS ξ(r)dr = E. − δ(β(r)) Solutions of these equations completely determine all the thermodynamic parameters and describe the general behavior of the selfgravitating system, whether this distribution of particles is spatially inhomogeneous or not. The above set of equations in principle solves the many-particle problem in the thermodynamic limiting case. The spatially inhomogeneous solution of these equations corresponds to the distribution of interacting particles. Such inhomogeneous behavior is associated with the nature and intensity of interaction. In other words, accumulation of particles in a finite spatial region (formation of a cluster) reflects the spatial distributions of the fields, activity, and temperature. It is very important to note that only in this approach, is it possible to take into account the inhomogeneous distribution of temperature that may depend on the spatial distribution of particles in the system. In other approaches, the dependence of temperature on the space point was introduced through the polytrophic dependence of temperature on the particle density in the equation of state [90]. In the present approach, such dependence follows from necessary thermodynamic conditions and may be determined for various distributions of particles. Now, we derive the saddle-point equation for the extremum of the local entropy function S(ϕ, ξ, β). The equation for the field variable δS/δϕ = 0 yields 1 Δϕ(r) + ξ(r) rm



2πm 2 β(r)

3/2



β(r) exp



 β(r)ϕ(r) = 0, (8.14)

where the notation rm ≡ 4πGm2 is introduced. The normalization condition may be written as 3/2     2m exp β(r)ϕ(r) dr = N (8.15) ξ(r) 2 β(r)

February 5, 2021

266

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 266

Statistical Physics of Nonequilibrium Systems

and the equation for the energy conservation in the system is given by       2πm 3/2 ξ(r)  3 3 − β(r)ϕ(r) exp β(r)ϕ(r) dr = E. 2 2 β(r) β(r) (8.16) To draw more information on the behavior of a self-gravitating system, we introduce new variables. The normalization condition  ρ(r)dr = N yields the definition for the density function, i.e.,    2πm 3/2 ξ(r) exp β(r)ϕ(r) , (8.17) ρ(r) ≡ 2 β(r) that reduces the equation to a simpler form. The equation for the field variable is given by  (8.18) Δϕ(r) + rm β(r)ρ(r) = 0. The equation for energy conservation takes the form    ρ(r)  1 3 − β(r)ϕ(r) dr = E. 2 β(r)

(8.19)

The equation thus obtained cannot be solved in the general case though it is possible to analyze some special cases of the behavior of a self-gravitating system under various external conditions. Hereinafter we write the chemical activity in terms of the chemical potential ξ(r) = exp(μ(r)β(r)). Having differentiated the equation for energy conservation over the volume, we obtain an interesting relation for the chemical potential, i.e.,  δE δV  1 ρ(r)  3 − β(r)ϕ(r) = = μ(r)ρ(r) (8.20) 2 β(r) δV δN that yields the chemical potential given by 3 1 β(r)ϕ(r). (8.21) μ(r)β(r) = − 2 2 Within the context of the expression for the density and the definition of the thermal de Broglie wavelength and the gravitation length, i.e.,  2m −1 , Rg (r) = 2πGm2 β(r) (8.22) Λ (r) = 2 β(r)

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Description of Nonequilibrium Systems

page 267

267

we can rewrite all the equations and the normalization condition in terms of density and temperature. Thus, we have   Rg (r) Λ3 (r)ρ(r)  + ρ(r) = 0 (8.23) Δ β(r) β(r) and the chemical potential reduces to 3 − ln(Λ(r)ρ(r)). (8.24) 2 In this approach, we can also obtain the equation of state for the self-gravitating system by using the thermodynamic relation μ(r)β(r) =

P =−

1 δS β δV

in the case of conservation of energy E. In our case, in the definition of “local entropy” we make use of the relation (∇ϕ(r))2 = ∇(ϕ(r)∇ϕ(r)) − ϕ(r)ϕ(r) and after that perform integration over the whole volume. The first part of the integration may be presented as a surface integral where ϕ(r) = 0 on the integration surface. After that, we can present the “local entropy” as    2πm 3/2 1 ϕ(r)Δϕ(r) − ξ(r) S= − 8πm2 G 2 β(r)   × exp β(r)ϕ(r) dr (8.25) and, using the definition of the particle density, rewrite it in the form    (8.26) S= −ρ(r) ln(Λ3 (r)ρ(r)) − ρ(r) dr. The local equation of state may be written as  1 3 . (8.27) P (r)β(r) = ρ(r)(1 − ln(Λ (r)ρ(r)) = ρ(r) μ(r)β(r) − 2 In the classical case Λ3 (r)ρ(r)  1 and P β ≡ ρ, but is logarithmically dependent on the particle density. Only in the case of Λ3 (r)ρ(r) = 1,

February 5, 2021

268

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 268

Statistical Physics of Nonequilibrium Systems

we obtain the usual equation of state for the ideal gas because in this case ϕ(r) = 0 and P β = ρ by virtue of nonoccurrence of interaction. For μ(r)β(r) = 3/2, the equation of state thus obtained reproduces the equation of state of the ideal gas. In this case, the energy of the system is equal to E = (3/2)N kT that is in accordance with the results obtained previously. In the next sections, we find the classical distributions of particles for various internal and external conditions. Homogeneous distribution of particles First of all, we consider the equilibrium case, all the parameters being independent of the space coordinates. In this case, both energy and total number of particles are fixed and, moreover, the temperature and the chemical potential do not change in space. Thus, the equation for the particle concentration   Rg (r) Λ3 (r)ρ(r)  + ρ(r) = 0 (8.28) Δ β(r) β(r) √ leads to a simple condition βρ(r) = 0 that can be realized only for T → ∞. The particle distribution in a self-gravitating system can be homogeneous only for very high temperatures. Another interesting case is when particle density depends on the coordinate while the temperature is fixed. In this case, the equation for the density takes the form   (8.29) Δ ln Λ3 ρ(r) + Rg ρ(r) = 0 and may be transformed to Δ (ln ρ(r)) + Rg ρ(r) = 0.

(8.30)

The latter equation has an exact solution ρ(r) =

2 , Rg r 2

but the normalization condition holds only for the case of a fixed box with size R = (N Gm2 /4kT ), fixed energy E = N kT , and the

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 269

Statistical Description of Nonequilibrium Systems

269

chemical potential density within the box is given by  2Λ3 3 − . μ = kT 2 4kT Rg r 2 As follows from the equation for constant temperature, the homogeneous distribution of particles is unstable. The homogeneous particle distribution ρ(r) = ρ + δρ(r) yields an equation for density fluctuations, i.e., Δδρ(r) + Rg ρδρ(r) = 0,

(8.31)

that reproduces the Helmholtz equation. The general solution of the with wave equation is the unstable radial distribution δρ(r) = exp(ikr) r  2 the wave number k = 2πGm βρ, that implies that the wavelength of the instability is half as long as the Jeans length. It is the statistical length of the instability of particle distribution in the system. The concept of Jeans gravitation instability is discussed within the framework of nonextensive statistics and is associated with the kinetic theory [118]. A simple analytical formula generalizing the Jeans criterion is derived by assuming that the unperturbed collisionless gas is kinetically described by the class of power-law velocity distributions. It is found that the critical values of the wavelength and mass depend explicitly on the nonextensive parameter. The instability condition is weakened as the system becomes unstable even for wavelengths of the disturbance smaller than the standard Jeans length. Recent discoveries of extrasolar giant planets, along with the refined models of the compositions of Jupiter and Saturn, suggest a reexamination of the theories of giant planet formation. An alternative to the favoured core accretion hypothesis is examined in [119], the conclusion is that the gravitation instability in the outer solar nebula leads to the formation of giant planets. Threedimensional hydrodynamic calculations predict the formation of locally isothermal or adiabatic thermodynamics. The gravitation instability appears to be able to form giant planets [120]. Our results can help to explain the data of astrophysical observations in the sense that the different length of the instability in a self-gravitating system

February 5, 2021

270

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 270

Statistical Physics of Nonequilibrium Systems

is associated with the alternative description of the situation. Thus we can conclude that particle distributions cannot be homogeneous for constant temperatures in the system. Hence, we have to find the true distributions of particles and temperature in the system. Inhomogeneous distributions of particles and temperature in a gravitating system In the general case, particle distributions in self-gravitating systems are inhomogeneous. Inhomogeneous distribution of particles gives rise to the long-range gravitation interaction. Now, we consider the nonequilibrium description of a self-gravitating system and take into account probable spatially inhomogeneous distributions of particles and temperature. We introduce a new variable ψ = Λ3 (r)ρ(r), then the equation for density is simplified, i.e., we have   Rg (r) ln ψ + ψ = 0. (8.32) Δ  β(r) β(r)Λ3 (r) The solution of this equation provides a completely nonequilibrium statistical description of a self-gravitating system. General exact solutions of this nonlinear equation are unknown. Further, we propose a way to solve this equation. First of all, we can find the most general solution of the problem. In the three-dimensional case the action of the Laplace operator may be presented in the form    2 1 d 2 d ln ψ  √ = ln ψ + Δ β dr 2 r dr β(r)    3 dβ 2 ln ψ d2 β 2 dβ − + − r dr 2β dr β 3 dr 2 1 d ln ψ dβ . − β 3 dr dr

(8.33)

One of the many exact solutions may be obtained if the second term in the right-hand part of this equation is equal to zero. Then the solution for the temperature is given by β = γ 3 r n where γ is an

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 271

Statistical Description of Nonequilibrium Systems

271

unknown constant and it is not difficult to see that for n = 2 we obtain only the equation for ψ, i.e., am d2 ln ψ ψ=0 + 2 dr Bγr

(8.34)

that may be rewritten in terms of the new variable r¯2 = r, i.e.,  4am d 1 dψ + ψ = 0. (8.35) d¯ r ψ d¯ r Bγ We multiply this equation by (1/ψ)(dψ/d¯ r ) and calculate the first integral of the equation thus obtained. It is given by  1 dψ 2 4am ψ=Δ (8.36) + ψ d¯ r Bγ and the exact solution may be written as ψ=

Δ  . (8am /Bγ) sinh2 (Δr/4)

(8.37)

Using the above definition, we find that the exact solution for the inhomogeneous distribution of particles is given by ρ(r) =

Δ 8am

γ 2 r 3 sinh2



(Δr/4)

(8.38)

and thus obtain good assumptions concerning the behavior at long distances from the centre of the inhomogeneous particle distribution. This behavior is related to the results obtained earlier in [114–116], where Boltzmann equation was used for the distribution function in the case of a spherical isolated stellar system. The distribution of particles is inhomogeneous for sizes R = 1/4Δ and divergent toward the centre, ρ(r) =

1 . 2am γ 2 r 4

In this case, the energy of the system is conserved. However, we do not know the coefficients. Thus, we propose the approach given below. If particles are concentrated at short distances and their concentration is very high, then quantum effects become a crucial

February 5, 2021

272

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Physics of Nonequilibrium Systems

factor and our approach is inapplicable. This is the degeneration condition. The relation between the critical temperature and particle concentration in this quantum case is determined by the natural condition  2 3/2  βc 3 ρc = 1. (8.39) Λ (r)ρ(r) = 2me According to this condition the quantum effect constrains the gravitation collapse in the system and determines the sizes of neutron stars. This relation along with the formula for the conservation of the number of particles, (4π)/(2am Rc ) = N determines all the required parameters, i.e., the critical distance Rc = (2 )/(mam N 1/3 ), the coefficient γ 2 = (2πme)/(2 N 2/3 ), the critical temperature βc = γ 2 Rc2 , and the concentration ρc = (1/(2am Rc4 ). The energy of the system is in this case given by E = (3/2)N kT , i.e., it is equal to the energy of a free particle. In this section, we present the general solution for the classical particle distribution at long distances from the centre of an inhomogeneous cluster of condensed matter that is subject to the laws of quantum physics. In all cases, our solution holds under the condition of classical physics, Λ3 (r)ρ(r)  1. In this section, we describe a system with Λ3 (r)ρ(r) = α = const  1. In this case, we can determine only the behavior of temperature that is governed by the equation   am eα 1 1 + = 0. (8.40) Δ  B ln α β β(r) Similarly to the previous case, the solution of this equation may be written in the form β = γ −2 r −2n and thus we find that it holds for n = −2, i.e., the temperature varies as kT = γ 2 r −4 , the concentration is varied as ρ = Ar −6 , and the normalization conditions for the conservation of particle number and energy are satisfied. The limiting behavior of the concentration and temperature provides a suitable solution to the problem in this special case. Finally, we make an attempt to present an arbitrary solution in the general case of spacecoordinate dependence of the concentration and temperature.

page 272

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Description of Nonequilibrium Systems

page 273

273

This equation describes any problem associated with inhomogeneous distributions of particles, temperature, and concentration in self-gravitating systems. Indeed, though the equation cannot be solved in the general case, it provides a possibility to analyze many cases of the behavior of a self-gravitating system under various external conditions. We may consider many realistic distributions of concentration, temperature, and field in a gravitation system but should not use the equation of state that exists as a condition related to this situation. In a realistic system, the temperature cannot be related to the particle distribution. It is a thermodynamic parameter that determines a condition for the behavior of the system and can be found from other physical reactions, not only gravitational. In the general case, we cannot obtain the general solution of the present equation, but can believe that this equation governs the thermodynamics of self-gravitating systems. We may conclude that the new approach in terms of the nonequilibrium statistical operator allowing for inhomogeneous distributions of particles and temperature is efficient. The statistical operator has no singularities for various values of the gravitation field. The approach makes it possible to solve the problem of selfgravitating systems of particles with inhomogeneous distributions of particles and temperature. The equation of state for the selfgravitating system has been determined. A new length of the statistical instability and parameters of the spatially inhomogeneous distribution of particles and temperature are found for realistic gravitational systems. The gravity factor can either promote or retard such transformations depending on the system and the conditions concerned. For the first time, a description is given of the formation of spatially inhomogeneous particle distributions accompanied by the changes of temperature. The statistical description of the system is tailored to treat the gravitating particles from an arbitrary spatial inhomogeneous particle distributions. In this approach, the probable behavior of a self-gravitating system can be predicted for any external conditions. Moreover, the method may also be applied for the further development of physics of self-gravitating and similar systems that are close to equilibrium.

February 5, 2021

274

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Physics of Nonequilibrium Systems

8.2 Systems with Repulsive Interaction The proposed approach is based on the nonequilibrium statistical operator [96] that is more suitable for the description of interacting systems. Next, a new approach is worked out to describe a system with purely repulsive interaction. The equation of state and all thermodynamic functions are determined by the equations that provide the dominant contribution to the partition function. We have also shown that this approach makes it possible to describe inhomogeneous distributions of particles and to determine the required parameters of the interacting system. The main idea is to provide a detailed treatment of an interacting system by the principles of nonequilibrium statistical mechanics and to show how to introduce the fundamental scalar field into the statistical description of the scattering in the system and thus to provide an illustration of the H-theorem. We have already mentioned that phenomenological thermodynamics is based on the conservation laws for the average values of physical parameters, i.e., the number of particles, energy, and momentum. Statistical thermodynamics of nonequilibrium systems is also based on the conservation laws, but for the dynamical variables rather than their average values. It represents local conservation laws for the dynamical variables. In order to determine the thermodynamic functions of a nonequilibrium system, a representation of the relevant statistical ensembles is needed allowing for the nonequilibrium states of the system. The concept of Gibbs ensemble may provide a description of nonequilibrium stationary states. In this case, we can determine a nonequilibrium ensemble as a set of systems contained under similar stationary external conditions. To determine a local equilibrium ensemble exactly, we have to find the distribution function or the statistical operator of the system [96]. We assume that nonequilibrium states of the system may be described by the inhomogeneous distribution energy H(r) and the microscopic  particle density n(r) = i δ(r−ri ). Starting from the nonequilibrium statistical operator of the local equilibrium distribution given by (8.1) with the Hamiltonian in the general case (8.3) for the repulsive

page 274

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 275

Statistical Description of Nonequilibrium Systems

275

interaction with ν = −1, the statistical operator [96] may be written in the form    p2 (r) β(r) Ql = DΓDψ × exp − 2m(r)    (8.41) − η(r) − i β(r)ψ(r) n(r)dr Qint , where Qint

   −1 1     β(r)U (r, r ) = exp − ψ(r)ψ(r )drdr 2

follows from the interaction written in terms of the field variable. The latter general functional integral can be integrated over the phase space. We substitute the definition expression for the density, then the nonequilibrium statistical operator reduces to       p2i − i β(ri )ψ(ri ) Qint , Ql = Dψ DΓξ(ri ) exp − β(ri ) 2mi (8.42) where ξ(r) ≡ exp η(r) is the chemical activity and should take into account that

1 dri dpi . DΓ = 3 (2π) N ! i Integration over all the moments reduces the real part of the nonequilibrium statistical operator to a simple expression [3, 8] given by         2πm(r) 3/2 ξ(r) cos β(r)ψ(r dr Ql = DψQint exp 3 β(r) (8.43) For the general case of long-range interaction, such as Coulomblike or Newtonian gravitation interaction in continuum, the limiting inverse operator should be treated in the operator sense, i.e., U −1 (r, r ) = −Lrr = −Lr δ(r − r ).

February 5, 2021

276

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 276

Statistical Physics of Nonequilibrium Systems

For the case of long-range interaction between particles, the nonequilibrium statistical operator may be written as    ψ(r)Lrr ψ(r ) Ql = Dψ exp −3

+ ξ(r)Λ

(r) cos



 β(r)ψ(r

 dr

(8.44)

where  Λ(r) =

2 β(r) 2m(r)

1/2

is the thermal de Broglie wavelength at each space point. In the general case, the nonequilibrium statistical operator is given by  (8.45) Ql = Dψ exp {S(ψ(r), ξ(r), β(r))} with the effective nonequilibrium “local entropy” being described by the expression     β(r)ψ(r dr. (8.46) S= ψ(r)Lrr ψ(r ) + ξ(r)Λ−3 (r) cos The statistical operator is suitable when applying the efficient methods developed in quantum field theory without additional restrictions on either integration over the field variables or the perturbation theory. The functional S(ψ(r), ξ(r), β(r)) depends on the field distribution, the chemical activity, and the reciprocal temperature. After that, we can apply the saddle-point method to find the asymptotic value of the statistical operator Ql as the number of particles N tends to infinity. The dominant contribution is given by the states that satisfy the extremum condition for the function. It is not difficult to show that the saddle-point equation represents the thermodynamic relation and may be reduced to an equation for the field variable, (δS)/(δψ(r)) = 0, the normalization condition  δS ξ(r)dr = N, δξ(r)

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 277

Statistical Description of Nonequilibrium Systems

277

and the energy conservation law for the system,  δS ξ(r)dr = E. δβ(r) The solution of this equation completely determines all the thermodynamic functions and describes the general behavior of interacting systems with both spatially homogeneous and inhomogeneous particle distributions. The above set of equations in principle solves the many-particle problem in the thermodynamical limiting case. The spatially inhomogeneous solution of these equations corresponds to the distribution of interacting particles. It is very important to note that only this approach makes it possible to take into account the inhomogeneous distribution of the temperature that depends on the spatial distribution of particles in the system.  From the normalization condition and definition ρ(r)dr = N , we may introduce some new variable that presents the macroscopic density function ρ(r) ≡ Λ−3 (r)ξ(r) cos( β(r)ψ(r)). In the case without interaction ϕ(r) = 0 for free particles, and if we write the chemical activity in terms of the chemical potential, ξ(r) = exp(μ(r)β(r)), then we obtain from this definition the well-known relation β(r)μ(r) = ln ρ(r)Λ3 (r) that generalizes the relation of equilibrium statistical mechanics [52, 53]. The equation for energy conservation in this case may be presented in the new form:     ρ(r)  1 3 − β(r)ψ(r)tan ( β(r)ψ(r)) dr = E. (8.47) 2 β(r) Derivation of the energy-conservation equation over the volume yields a relation for the chemical potential, i.e.,  δE δV   1 ρ(r)  3 − β(r)ψ(r)tan ( β(r)ψ(r)) = = μ(r)ρ(r) 2 β(r) δV δN (8.48) hence the chemical potential is given by μ(r)β(r) =

  3 1 − β(r)ψ(r)tan β(r)ψ(r) . 2 2

(8.49)

February 5, 2021

278

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Physics of Nonequilibrium Systems

This approach also provides the equation of state for the system within the context of the thermodynamic relation for pressure P = (1/β)(δS/δV ) in the case of energy conservation. The local equation of state is now reduced to  1 . (8.50) P (r)β(r) = ρ(r) μ(r)β(r) − 2 In the case of an ideal gas, ψ(r) = 0, we have μβ = 3/2 and obtain the usual equation of state P β = ρ that reproduces the equation of state of the ideal gas. The energy of the system is equal to E = (3/2)N kT , this formula is in accordance with the previous well-known results [53]. Within the context of the definition (8.50) we thus conclude that, under the condition μ(r)β(r) < 1/2, there arises negative pressure P (r) < 0 that satisfies the necessary vacuum condition in the  cosmology. It holds under the special condition  β(r)ψ(r)tan ( β(r)ψr) < 2 for constant temperature and for the total energy of the system E < (1/2)N kT . This condition implies that the energy of each particle is lower than the thermal energy. In this special case, the energy of the system is lower than the total thermal energy of particles, that is impossible. The saddlepoint method provides a possibility to select system states whose contributions in the partition function are dominant [1, 3]. The solutions of the equations thus obtained are associated with finite values of the functional and may be regarded as thermodynamically stable particle and field distributions. With these solutions we have to determine the general relation between thermodynamic parameters and their spatial dependence. Thus the spatially inhomogeneous distribution of fields can be unambiguously related to the spatially inhomogeneous particle distribution. In the general approach, all thermodynamic parameters (pressure, chemical potential, density) depend on the space point and are mutually dependent. Actually, this approach extends the mean-field approximation to involve spatially inhomogeneous field distributions. In our case, the field that is introduced describes the nature of the repulsive interaction and can present the fundamental scalar field that corresponds to the scattering in the system.

page 278

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 279

Statistical Description of Nonequilibrium Systems

279

 We introduce a new field variable ϕ = β(r)ψ(r) and take into account the definition of the chemical potential (8.49). Then, we may rewrite the macroscopic density in the form given by   1 −3 (8.51) ρ(r) ≡ Λe (r) exp − ϕ(r)tan ϕ(r) cos ϕ(r), 2 where the de Broglie wavelength is renormalized, i.e., 1/2  2  β(r)e . Λe = 2m(r) After substituting the relation thus obtained in (8.46) we find that the local entropy in terms of the new variable is given by   ϕ(r ) ϕ(r)  Lrr  + Λ−3 S= e (r) β(r) β(r)  

1 (8.52) × exp − ϕ(r)tan ϕ(r) cos ϕ(r) dr. 2 For constant temperature and equal particle masses the local entropy in the mean-field approximation may be presented as   1 ϕ(r)Lrr ϕ(r ) + Λ−3 S= e β  

1 (8.53) × exp − ϕ(r)tan ϕ(r) cos ϕ(r) dr. 2 Now the equation for the field variable may be rewritten as 2Lrr ϕ(r ) − β where the potential energy V (ϕ) ≡ ρ(r) =

Λ−3 e exp

dV (ϕ) = 0, dϕ

 1 − ϕ(r)tan ϕ(r) cos ϕ(r) 2

(8.54)

(8.55)

is a function of the field variable in the above form. This potential has a minimum for 3 sin 2ϕ = −2ϕ. For small values of ϕ, we have two different solutions, ϕ = 0 and ϕ2 = 1. For small ϕ, the effective

February 5, 2021

280

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 280

Statistical Physics of Nonequilibrium Systems

potential is given by a very simple expression V (ϕ) = (1 − ϕ2 ) and the equation for the field variable reduces to 2Lrr ϕ(r )+2βϕ(r ) = 0. In the general case, the potential energy of the field possesses oscillatory character with decreasing amplitude. This representation with the necessary condition of the cosmological model of natural inflation was proposed in [238]. After that, we can analyze the probable spatial solution for the field variable and the behavior of the field in time. In order to provide this, the knowledge of particle interaction energy with relevant specifics is required. It is wellknown from cosmology reasoning that galaxy scattering is associated with the fundamental scalar field. We have shown above that the repulsive statistical interaction can be described by the field ϕ =  β(r)ψ(r). We suppose that the introduced fields are fundamental scalar fields responsible for the statistical motivation of the scattering of matter. From this assumption, we find the energy of interaction between two masses located at different space points. As follows from cosmology, two masses scatter with the velocity v = H(r − r ), where H is the Hubble constant and r − r is the distance between them. The kinetic energy of the relative motion for each mass is given by T =

m(r) 2 H (r − r )2 2

and thus the energy of interaction between two masses located at different space points is W (r − r ) =

m(r) 2 m(r ) 2 H (r − r )2 + H (r − r )2 . 2 2

For homogeneous distribution of masses, the latter expression may be rewritten as W (r−r ) = m(r)H 2 (r−r )2 . In terms of such interaction energy, the inverse operator is given by Lrr =

1 d2 . mH 2 dr 2

(8.56)

Having determined the inverse operator, we can present the spatial dependence of the fundamental scalar field as the solution of the

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 281

Statistical Description of Nonequilibrium Systems

281

equation dV (ϕ) 2 d2 ϕ = 0. −β mH 2 dr 2 dϕ

(8.57)

For small values of ϕ, the latter transforms to the equation d2 ϕ + βmH 2 ϕ = 0, (8.58) dr 2 √ that has a periodic solution ϕ = cos( mβHr) with the spatial period √ (1/ mβH). For distances shorter than this value, we may regard the fundamental scalar field to be invariable. However, the fundamental scalar field can change in time. To describe the evolution of such fields, we should present a dynamical equation. In our case, however, this field induces the repulsive interaction in the system and causes entropy increase. The behavior of the solution of the equation is similar to the behavior that follows from the usual equation for scalar field and the formation of a new-phase bubble of the fundamental scalar field. The present solution can describe the formation of a bubble of a new phase in the theory of inflation of the Universe [24, 78], and the field variable introduced plays the role of the fundamental scalar field and takes into account the repulsive interaction in the system under consideration. In the general presentation, formula (8.46) can describe the condition of new phase formation, the size of the bubble, and other parameters of the thermodynamic behavior of such systems. This nonequilibrium statistical describes only probable dilute structures of such systems, it does not describe metastable states and tells nothing about the time scales in the dynamical theory. In this way, however, we can solve complicated problems of the statistical description of interacting systems. For this purpose, we have to derive a dynamical equation for the field. In this sense, we can use the Ginsburg–Landau equation for the fundamental scalar field in the standard form given by δS ∂ϕ(r, t) = −γ , ∂t δϕ(r)

(8.59)

February 5, 2021

282

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Physics of Nonequilibrium Systems

where γ is the dynamical viscosity coefficient [65]. In this case, all the necessary conditions satisfy the thermodynamic relation. We may suppose that the motivation of the Universe dynamics is associated with the entropy increase. The evolution in the nonequilibrium state is governed by the local entropy landscape and the morphological instabilities of the parameter. The dynamics of the system is dissipative, it should result in the decrease of the local entropy. This solution of the obtained equation provides the answer to the question, what is the motive of the scattering of matter. The dynamics of the formed Universe can be only dissipative and in order to describe the existence of the Universe we have to take into account its nonequilibrium conditions. Interacting particle systems are nonequilibrium a priori. Before relaxing into thermodynamic equilibrium, isolated systems with longrange interactions are trapped in nonequilibrium quasi-stationary states whose lifetimes diverge as the number of particles increases. A theory that makes it possible to quantitatively predict the instability threshold for spontaneous symmetry breaking in a class of d-dimensional systems was proposed in [94]. Nonequilibrium stationary states were described in [95]. The authors have drawn a conclusion that three-dimensional systems do not evolve to thermodynamic equilibrium but are trapped in nonequilibrium quasistationary states. We propose an approach that provides a possibility to quantitatively predict the particle distribution in a system with special repulsive interaction. In this way, we can solve the complicated problem of the statistical description of systems with special repulsive interactions and introduce a new field variable that reduces this task to the solution of the cosmological problem. Moreover, this method may also be applied for the further development of physics of self-gravitating and similar systems that are not far from equilibrium.

8.3 Saddle States of Nonequilibrium Systems In all cases, a macroscopic system that interacts with the environment turns into the equilibrium state after the relaxation time.

page 282

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Description of Nonequilibrium Systems

page 283

283

The properties of such systems are determined by the specifics of each system and characteristics of the environment. The equilibrium state of a separate macrosystem may be determined under ideal conditions [53, 65, 117]. Due to the influence of the environment, the thermodynamic parameters of a separate macrosystem become equal to those of the thermal bath. Within the context of the fundamental principles of thermodynamics, any macroscopic system embedded in a thermal bath approaches equilibrium during some relaxation time. In the equilibrium state, the properties of the system do not depend on how the equilibrium has been established. The equilibrium state is, however, realized only under certain idealized conditions, so in reality, the properties of the system in a quasi-stationary (steady) state may depend both on the specifics of the interaction of the system with the thermal bath and the characteristics of the bath [53–65]. The same concerns nonequilibrium systems. In the case of a nonequilibrium system the combined parameter cannot be determined. Nevertheless, such systems may possess equilibrium behavior. As is shown below, it is possible to determine the equilibrium state of the system as a stationary state for which the energy interchange between the separate system and the environment is balanced. The state of the system is the result of the balance between the direct influence of the environment on the system and the degradation process caused by the interaction with the environment. It is not difficult to imagine a macroscopic system that can receive energy from the environment but cannot return it back. This depends on the specifics of the system. Examples of such systems are thermal electrons in semiconductors [96]; a system of photons that diffract from inhomogeneities with the diffraction coefficient depending on the frequency of these photons [239, 240]; a system of high-energy particles that may be born in the course of particle collisions in an accelerator; a system of ordinary Brownian particles, when the friction coefficient depends on the velocity. All these systems exist far from thermal equilibrium and the new state is completely determined by the processes of energy exchange. These systems may be described by the distribution functions of the states

February 5, 2021

284

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Physics of Nonequilibrium Systems

that may differ from the well-known distribution functions of thermal states. There are no well-defined methods for determining nonequilibrium distribution functions that would take into account probable states of the macroscopic system [96]. The standard method makes it possible to describe a nonequilibrium state basing on the information on the equilibrium thermal state and small deviations from this state. The nonequilibrium is in this approach manifest as slight modifications of the equilibrium distribution function. An open system can exist very far from equilibrium but nevertheless manifest stationary behavior. In this section, we consider a certain problem related to the description of a nonequilibrium system and analyze a probable definition of a new stationary state taking into account energy dissipation or absorption along with the degradation processes caused by the interaction with the environment. It is well-known that any state of a system may be described in terms of distribution functions that determine all thermodynamic properties of macroscopic systems [53, 117]. Actually, the statistical description of a macroscopic system requires the knowledge of only few macroscopic parameters, e.g., energy. Hence, it is a fundamental task to work out a method for the study of the general properties of steady states of open systems and to reveal the conditions for such states to exist. One of the probable ways to solve this general problem may be based on the Gibbs approach [241]. The main purpose is to propose a simple way of describing nonequilibrium systems in the energy space [158] and to formulate a new concept of the solution of the cosmological problem. We start from the formulation of the statistical approach. The canonical partition function in phase space in the equilibrium case may be written as   F − H(q, p) dΓ, (8.60) ρ(q, p)dΓ = exp Θ where H(q, p) = ε is the Hamiltonian on the hypersurface of constant energy ε, dΓ = i dqi dpi is an element of the phase space, Θ = kT , T is the temperature, and F is the free energy that may be found

page 284

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Description of Nonequilibrium Systems

page 285

285

from the normalization condition    F − H(q, p) dΓ = 1. exp Θ As it is shown in [241], the phase space depends only on energy and external parameters. We introduce an additional function Σ = ln(dΓ)/(dE), then the canonical partition function reduces to the form given by   F −ε + Σ(ε) dε. (8.61) ρ(ε)dε = C exp Θ The latter presentation makes it possible to describe the dependence of the distribution function on the energy of the macroscopic system [241]. The normalization condition in this presentation may be written in the form    F −ε + Σ(ε) dε = 1 (8.62) c exp Θ from which one can find the normalization constant allowing the determinant transformation between the phase space and the energy variable. In order to select the states that most contribute to the partition function we employ the condition (dΣ)/(dε) = 1/Θ that determines the temperature of the system provided the change of phase space as a function of the system energy is known. Using this definition and taking into account the basic principles of statistical mechanics [65], we come to the conclusion that Σ = ln(dΓ)/(dε) = S is equal to the entropy of the system. Now, we have to make a very important note: the temperature describes the dependence of the entropy only on energy, but not on any other thermodynamic quantities. We can define the temperature for other situations, but this definition makes no sense without changing the entropy. Another important conclusion is that we can calculate the partition function by integrating over energy. Such integration in this sense implies a continual integral over the energy variable. The extremum of the partition function is realized under the condition F = E − θS and any probable deviation from this condition makes a very small contribution to the macroscopic characteristics similarly to the

February 5, 2021

286

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 286

Statistical Physics of Nonequilibrium Systems

quantum contribution to classical trajectories. The novelty of this presentation consists in the possibility to consider the nonequilibrium systems as Brownian systems in the energy space [158]. Starting from the basic kinetic equations for the distribution function of the macroscopic system in the energy space, we can obtain steady states and fluctuation dissipation relations for the nonequilibrium systems [158]. The energy, as a control parameter of the nonequilibrium system, can be a “slow” parameter that determines the state of the system. In the absence of any other knowledge about the nonequilibrium system, there is no reason to prefer any state of the system determined through energy. The system energy variation determines the state of the system. The nonequilibrium distribution function, similarly to the equilibrium case, may be defined as ρ(ε, t), which includes the dependence on the energy of the system ε and time. The energy distribution function, in the general case, may be obtained from the basic kinetic equation that presents the evolution of the system during a long period of time and takes into account probable fast processes that might occur in it. In terms of the energy, the basic kinetic equation for the nonequilibrium distribution function may be written as   ∂ρ(ε, t) = W (ε, ε )ρ(ε , t) − W (ε , ε)ρ(ε, t) dε , (8.63) ∂t where W (ε, ε ) is the probability of a transition between different energies of the system. This basic kinetic equation represents the balance equation for the probabilities of states. The energy presentation of a nonequilibrium process is valid only in the case when this variable is canonical and it is possible to carry out averaging. All solutions of the basic kinetic equation for t → ∞ possess a fundamental property — they reduce to the stationary solution that may be interpreted as an “equilibrium” solution for this system. This stationary solution complies with the H-theorem [242]. In the special case, the basic kinetic equation reduces to the Fokker–Planck equation ∂ 1 ∂2 ∂ρ(E, t) = A(ε)ρ(ε, t) + D(ε)ρ(ε, t) ∂t ∂ε 2 ∂ε2

(8.64)

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 287

Statistical Description of Nonequilibrium Systems

287

that may be derived from the basic kinetic equation in the approximation of fast changes of the energy growth rate and slow changes of the distribution function as a function of the control parameter. The coefficients A(ε) and D(ε) depend on energy and their physical meaning should be specified for various situations. The physical meaning of the coefficients may be clarified when we come back to the dynamic equation for the energy. In the general case, one may suppose that the dissipation equation in the standard form is given by dε = f (ε) + g(ε)L(t). dt

(8.65)

This dissipation equation depends on the external influence and initial conditions. The external influence, first of all, is manifest in the change of the energy of the system that dissipates or is absorbed due to the external action. This process is taken into account by the first part of the present equation that describes the direct influence of the environment on a separate macroscopic system. This part can be obtained from the dynamics of any macroscopic system provided the direct interaction of this system with the environment is completely determined. But this is true not for all cases. In the general case, however, the system exists in a contact with the nonlinear environment. The system energy is changed due to the random influence of the environment and this phenomenon is taken into account by the second part of the equation. The random migration of the system is the result of interaction between this system and all other macroscopic systems, the influence of which randomly changes the energy of the system. The influence of this interaction is not correlated and the correlation between two values of fluctuations at two different moments L(t)L(t ) = φ(t − t ) can be nonzero only for the time interval that is equal to the time of action. The symbol . . . implies the statistical averaging of the a drastic peak in relevant value. The function φδ(t − t ) must have  the vicinity of zero and satisfy the condition φ(τ )dτ = σ 2 for the white noise [242]. The system that cannot come to equilibrium after fast changes of the environment should relax to a new state. This process suggests probable degradation of the system in contact with

February 5, 2021

288

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 288

Statistical Physics of Nonequilibrium Systems

the environment. The energy presentation of the general dissipation equation takes place also in the case of ordinary Brownian particles. The dynamics of Brownian particles may be described in terms of the velocity v by the Langevin equation dv = −γv + F (t), dt

(8.66)

where γ is the friction coefficient and F (t) is the random force associated with the action of the fluid on a particle under the condition that the average over the equilibrium ensemble vanishes, F (t) = 0, and F (t)F (t ) = φ2 δ(t−t ) which satisfies the condition for white noise and describes uncorrelated motion of particles. For a Brownian particle with the energy ε = M v 2 /2 the energy change may be written as √ dv dε = Mv = −2γε + 2M εF (t), dt dt

(8.67)

√ that reproduces √ the previous equation with f (ε) = −2γε, g(ε) = ε and L(t) = 2M F (t). Using the solution of the Langevin equation for the velocity, we obtain [242] v(∞) = (φ2 /2γ) = (kT )/M and so ε = (kT )/2, where the temperature of the thermal bath T is introduced. As a solution of the equation in the approximation disregarding the correlation energy, one can also obtain 

ε2 =

φ2 σ2 ≡ 2M = kT, 4γ 4γ

that, similarly to the previous result, completely satisfies the equilibrium condition. Different descriptions of the processes occurring in a nonequilibrium system are equivalent, however, the energy presentation is preferable because it facilitates the determination of the condition for the “equilibrium” states of a nonequilibrium system. This approach is valid for various systems for which the direct influence of the interaction with the environment and probable random nonequilibrium fluctuations can be determined. It is preferable because the energy is the slowest variable on which depends the relaxation of the system.

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 289

Statistical Description of Nonequilibrium Systems

289

The true nonlinear Langevin equation should have an equivalent equation for the probability distribution function that considers the relevant physical process. Up to now, two different approaches have been proposed. The first one assumes that the coefficient g(ε) depends on the energy at the starting point, thus the equation for the nonequilibrium distribution function may be obtained in the Ito form. If this coefficient depends on the energy before and after the transition, then the diffusion equation may be written in the Stratonovich form, i.e.,  σ2 ∂ ∂  ∂ ∂ρ =− f (ε)ρ + g(ε) g(ε)ρ. ∂t ∂ε 2 ∂ε ∂ε

(8.68)

In what follows, we employ only Stratonovich presentation, because both presentations can be used [76, 242]. Both equations, in different forms, do not make much sense. The physical interpretation of the processes should be studied. In the case under consideration, different states of any system are determined and can be formed by the previous state and probable future states. The present equation for the nonequilibrium distribution function in such a case may be rewritten in more usual form of the local conservation law for the probability, i.e., ∂J(ρ(ε, t)) ∂ρ(E, t) = . ∂t ∂ε The flow of probability may be written as 

∂ σ2 ∂ σ2 J = − f (ε) − g(ε) g(ε) ρ + g2 (ε) ρ. 2 ∂ε 2 ∂ε

(8.69)

(8.70)

The comparison of both Fokker–Planck equations in the energy presentation yields A(ε) = f (ε) −

∂ σ2 g(ε) g(ε) 2 ∂ε

and the diffusion coefficient D(ε) =

σ2 2 g (ε). 2

February 5, 2021

290

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 290

Statistical Physics of Nonequilibrium Systems

The stationary solution of the Fokker–Planck equation for J(ρ(ε, t)) = 0, without flow of probability through each energy point, may be given by   ε g(ε) 2f (ε )dε − ln . (8.71) ρs (ε) = A exp 2 2  g(ε0 ) ε0 σ g (ε ) The equilibrium distribution function as a stationary solution in the general nonequilibrium case may be presented as ρs (ε) = A exp {−U (ε)} , where g(ε) − U (ε) = ln g(ε0 )



ε ε0

2f (ε )dε . σ 2 g2 (ε )

(8.72)

(8.73)

This distribution function has the extremum value of energy that may be found as a solution of the equation  1   D (ε) − f (ε) , ε) = (8.74) U  (! D(ε) where  stands for the energy derivative. This equation is equivalent to the equation ε) = f (! ε), D  (!

(8.75)

that determines the relation between the dissipation in the system and the diffusion in the stationary case and thus completely determines the new “equilibrium” state of the system. The stationary nonequilibrium distribution function is given by   ε)} exp −U  (! ε)ε2 , (8.76) ρs (ε) = exp {−U (! where ε) = −U  (!

 1   D (! ε) − f  (! ε) . D(! ε)

This “equilibrium” distribution function is of Gaussian form. When the dissipation f (ε) is represented by the nonlinear function of the state, and the diffusion coefficient depends on the energy, many interesting situations, including the noise-induced transition into new,

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Description of Nonequilibrium Systems

page 291

291

more stable, “equilibrium” states, may be obtained. The probable cases that may be described in this approach are given below. (a) In the case when the diffusion coefficient g(ε) = 1, irrespective of energy, the stationary solution may be written in the form   ε f (ε)  dε , (8.77) ρ(ε) = A exp 2 ε0 σ where ε0 is then intrinsic energy of the system. In the case of a conservative system, for f (ε) = 0, the stationary solution transforms into constant. It should be noted that ε = ε0 is not only the intrinsic limit but also the stationary point under the absence of the dissipation energy and random diffusion. Moreover, this point is the attractive point and thus the whole stationary distribution function should have an extremum according to the normalization condition for the distribution function [76]. Only in this case, the distribution function can reproduce the microcanonical distribution function. (b) Furthermore, in the case when only random diffusion of energy occurs, for g(ε) = 1, the equation for the nonequilibrium distribution function takes the form of the simple diffusion equation, the solution is given by   (ε − ε0 )2 1 exp − ρ(ε) = A √ 4σ 2 t 4πσ 2 t and describes the migration of the system over the energy. The measure of blueness increases in time according to the law (ε − ε0 )2  = 2σ 2 t. This solution represents the evolution of the system that in the initial state is described by the equilibrium distribution function ρ(ε) = δ(ε − ε0 ). All states of the system at the initial time are on the surface of constant energy. The fluctuations of the external medium are manifest in the absence of the microcanonical distribution and the occurrence of the uniform distribution over any energy. (c) If the diffusion coefficient depends on energy and the energy conservation law holds, f (ε) = 0, then the stationary solution is

February 5, 2021

292

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 292

Statistical Physics of Nonequilibrium Systems

given by   g(ε) ρs (ε) = A exp − ln g(ε0 )

(8.78)

that corresponds to the canonical equilibrium distribution function ρ(ε) = A exp {−β(ε − ε0 )}

(8.79)

only if g(ε) = eβε , where β may be presented as the reciprocal temperature of the environment. The latter relation holds only if the diffusion coefficient takes the form D(ε) =

σ2 2 σ 2 2βε g (ε) = e . 2 2

The physical conditions and specifics of the interaction of the system with the external medium should allow for the degradation processes in the system. (d) When the presentation of the dissipation energy in the system is f (ε) = 0 with regard to the previous presentation of the diffusion coefficient, the equilibrium distribution function may be written in the form   

f  (! ε) 2 ε . (8.80) ρs (ε) = A exp −2β 2β − f (! ε) The diffusion coefficient is a universal characteristic of the environment. The stationary solution for the distribution function can be realized only for special relations between the induced and degradation energies. If the system gets as much energy as possible through the contact with the environment, then the stationary state will not be observed. The energy presentation may be more illustrative for the ordinary description of equilibrium states. For example, for an ordinary Brownian particle, the stationary solution may be written in the form   √ 4γ 1 (8.81) ρs (ε) = A exp − 2 ε − ln ε ≡ A √ exp(−βε), σ ε

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 293

Statistical Description of Nonequilibrium Systems

293

2 where we have applied the well-known relation  (2γ)/(σ ) = β. Taking into account the normalization condition ρs (ε)dε ≡ ρs (p)dp, we obtain the equilibrium distribution function in the momentum space as given by     M v2 p2 = A exp − . (8.82) ρs (p) = A exp −β 2M 2kT

The stationary solution completely represents the well-known equilibrium distribution function for ordinary Brownian particles. The energy dependence of the diffusion coefficient may also be obtained from the linear Langevin equation by using the theory of Markovian processes and taking into account the nonequilibrium fluctuations of various coefficients in the function f (ε). If the dissipation function may be presented in the form f (ε) = αt eβε , then the dissipation Langevin equation may be rewritten in another form, i.e., de−βε(t) = −βαt , (8.83) dt where αt = α + ξt contains a constant part and the part ξt that describes the influence of the environmental white noise [76]. The Fokker–Planck equation for the nonequilibrium distribution function may be presented in terms of the new variable z = e−βε in a simple form given by  σ2 β 2 ∂ 2 ∂  ∂ρ(z, t) = αβρ(z, t) + ρ(z, t). ∂t ∂z 2 ∂z 2 The stationary solution   2α z ρs (z) = exp σ2β

(8.84)

in the case βε > 1 may be presented as ρs (ε) = exp{−βε} for (2α)/(σ 2 β) = 1. This solution is equal to the solution obtained previously by the standard approach. In this approach, we can also consider a very simple picture of the motion of Browning particles in a heterogeneous medium. For this matter the characteristic of the medium may be taken into account by

February 5, 2021

294

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 294

Statistical Physics of Nonequilibrium Systems

different values of the friction coefficient. For example, the motion of a large particle in a suspension, a dust particle in a nonhomogeneous medium, and some other interesting systems in the cases when it is necessary to determine the kinetic properties of nonlinear particlevelocity dependence. Therefore, in the above Langevin equation we have to take into account that γt = γ +ξt consists of a constant part γ that determines the average friction coefficient, and the chaotic part ξt that describes the influence of random changes in the friction of matter. In the case of white noise, the Fokker–Planck equation may be written as [76]  σ2 ∂ 2 2 ∂ ∂ρ(ε, t) = γερ(ε, t) + ε ρ(v, t). ∂t ∂ε 2 ∂ε2 The stationary solution of this equation is [76] − 12



ρs (ε, t) = N ε

(8.85)



2γ +1 σ2

,

(8.86)

which may be verified by the direct substitution of the solution in the previous equation. This stationary solution differs from the solution in the standard case, when both diffusion and friction coefficients do not depend on energy. A similar result ρs (v, t) = N v

−





2γ +1 σ2

may also be obtained in the velocity presentation. It is possible to present the situation when the system energy increases but there exists a mechanism that limits the energy. For a process that may be described in terms of the dissipation function f (ε) = γε − ε2 , the second part just takes into account such limiting. The absorption parameter may be presented as γt = γ + ξt where the second part describes the random change of the environmental influence. The Fokker–Planck equation takes the form given by [76]  σ2 ∂ 2 2 ∂ ∂ρ(ε, t) = (γε − ε2 )ρ(ε, t) + ε ρ(v, t). ∂t ∂ε 2 ∂ε2 The stationary solution of this equation may be written as     2 − 2γ2 +1 σ exp − 2 ε ρs (ε, t) = N ε σ

(8.87)

(8.88)

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Description of Nonequilibrium Systems

page 295

295

Fig. 8.1. Dependence of the velocity distribution function of charged grains on the charge of moving grain [4, 38, 62].

which resembles though does not reproduce the thermal distribution function. Illustration of this approach describes the behavior of dusty particles in weekly ionized plasma as seen in Fig. 8.1. A probable approach to the description of nonequilibrium states is proposed. The Fokker–Planck equation for the nonequilibrium distribution function of a macroscopic system is employed to obtain a stationary solution that may be interpreted as the “equilibrium” distribution function for the new energy state. The approach takes into account probable transitions between various states of the system induced by the energy dissipation and influence of the environment that depends on the energy of the system. Nonlinear models are described that represent probable stationary states of a system with various spatial processes in it. 8.4 Nonequilibrium Dynamics of Universe Formation Standard cosmological models involve a scenario of Universe nucleation and expansion based on the scalar field that is of fundamental importance for the unified theories of weak, strong, and electromagnetic interactions with spontaneous symmetry breaking. An important unexpected feature of the field theory with spontaneous symmetry breaking is that the Universe lifetime in a metastable energydisadvantageous vacuum state of this field is very long. A theory of new-phase bubble nucleation and expansion was proposed in [243] and extended in [24, 244]. Various cosmological models describe

February 5, 2021

296

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Physics of Nonequilibrium Systems

tunneling through the potential barrier in terms of the potentials V (ϕ) of arbitrary forms. At the same time, the cosmological effect of the scalar fields has been proposed as a mechanism to drive the evolution of the Universe in various scenarios. A probable way to solve the fundamental problem in standard cosmology is to consider the inflation Universe of present times. The above-mentioned model takes into account the interaction of a fundamental scalar field with probable fluctuations of fields of other nature. Such fluctuations may be described as the multiplicative vacuum noise. In this case, multiplicative noise not only changes the value of a scalar field, but also modifies the form of the effective potential that modifies the state of the system. This in turn changes the conditions of new phase bubble formation and the evolution of the Universe. The present model differs from the widely studied scenario of the stochastic inflation of the Universe [24, 78] that does not take into account external fluctuations of the fundamental field but allows for the internal field fluctuations of the unstable environment. More realistic models should consider both probable fluctuations as the fundamental scalar field and fields of other nature that arise from the initial condition. The external fluctuation leads to the disorganizing action. The stationary state of the system depends on the adjustments of the potential fluctuations of the environment and the stationary behavior of the system changes [76]. The internal fluctuations change the profile of the potential. However, in order to calculate the size of the bubble we have to violate the equivalence of the local minima. The degree of violation enters the final formula for tunneling probability. The external fluctuations deviate the system from the equilibrium due to intensive interactions of the scalar field with probable fluctuations of the fields of other nature. In this case, the probability to find the system in a given position is presented by the distribution function. The dynamics of the Universe may be described as a stochastic process with a new stationary state formation in the fundamental scalar field that determines the necessary cosmological parameters. In this sense, the new phase can be formed for arbitrary values of the scalar field while the dispersion of fluctuations fully determines the conditions of new phase bubble formation.

page 296

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 297

Statistical Description of Nonequilibrium Systems

297

A cosmological phenomenon for which fluctuations are crucial is transient dynamics associated with the relaxation from states that have lost their global stability due to the change of appropriate control parameters that in our case is the fundamental scalar field. No unified commonly accepted theoretical approaches have been proposed to determine the probable states of the vacuum. Such systems, however, are abundant in nature and it is a fundamentally important task to develop a method to explore the general properties of the stationary states of the vacuum and to establish the conditions of their existence. A simple model of noise-induced formation of the Universe and its dynamics that allows for the inflation at the present time, is being proposed here. For the sake of definiteness, we consider the decay of vacuum with different stationary states by the theory of fundamental scalar field that employs the Lagrangian given by L=

1 (∂μ ϕ)2 − V (ϕ) , 2

(8.89)

where model potential can be presented in the well-known form: V (ϕ) = ε −

μ2 2 λ 4 ϕ + ϕ , 2 2

(8.90)

where μ2 is the mass coefficient, λ is the coupling constant. The minimum of the potential in the regime of spontaneous symmetry breaking is characterized by ϕ20 =

μ2 , λ

V (ϕ0 ) = ε −

μ4 . 4λ

In particular, the static Lorentz-invariant Universe requires that the parameter of the potential should be related by ε = (μ4 )/(4λ). In the limiting case that ignores the matter and radiation but maintains the homogeneity and isotropy of the Universe, the dynamics of the system is governed by the equation [78, 245] 3H ϕ˙ = −

dV (ϕ) , dϕ

(8.91)

February 5, 2021

298

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 298

Statistical Physics of Nonequilibrium Systems

where ϕ˙ denotes time variation and H2 =

8πGV (ϕ) 3

(8.92)

determines theHubble parameter that for ϕ = 0 is constant, (8πGε)/3. If we introduce a new time variable H = H0 = τ ≡ t/3H0 , then the previous dynamic equation for the constant Hubble parameter may be rewritten as ϕ˙ = μ2 ϕ − λϕ3 .

(8.93)

Now, we suppose that the fundamental scalar field does not completely determine the state of this nonequilibrium system. Such system fluctuations of any other field existing can play an important role in the formation of stationary state. Particularly, fundamental scalar field can interact with such fluctuations that leads to the changes of the state. Moreover, it is shown in [76] that the noise induces a phase transition into the new state and determines the lifetime in the metastable state [246]. The time evolution of the uniform scalar field with regard to the noise is described by the stochastic equation dϕ = μ2 ϕ − λϕ3 + σξ. dt

(8.94)

In this case, the relaxation √ dynamics of the scalar field depends on the parameter k = μ2 / λσ that is included in the expressions for the lifetime of the state ϕ = 0 and the time dependence of the distribution function P (x, t) [246]. For small noise intensity σ 2 , the mean lifetime in the metastable state ϕ = 0 may be written as [246] " " " 1 1 μ2 "" " , (8.95) T (0 → ϕ0 ) = √ Φ(k) + 2 ln "1 + 2μ 4λϕ0 " λσ where Φ(k) and may be presented by the zero-order modified Bessel function, i.e.,  ∞  kn 1 ∞ −ku 2 (8.96) e K0 (u )du = (−1)n Bn , Φ(k) = 2 0 n! n=0

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 299

Statistical Description of Nonequilibrium Systems

where

299

√



 2 n/2 n+1 2 . 2 Γ Bn = 16 4

If k 1 and the barrier height ΔV = μ4 /4λ, then we have Φ(k) ≈ (1/2k) ln k 2 and the mean lifetime yields the value T  = (1/2μ2 ) ln(aϕ0 /σ). This description is related to the scenario of the interaction of the fundamental scalar field and white noise fluctuations. In nonequilibrium physics, among the phenomena where the effect of statistical fluctuations is crucial, is transient dynamics associated with the relaxation of states that have lost their global stability due to the changes in appropriate control parameter which in the proposed stochastic model is the scalar field. In this sense, we may change the state of the system and change all the parameters that determine the nonequilibrium dynamics of the system. Nevertheless, it is possible to define the new stationary states for such nonequilibrium systems, whose distribution function is different from the well-known equilibrium distributions. Generally, we may assume that the equation that describes the changes of the scalar field in such a nonequilibrium system as the Universe with various fluctuations has the form equivalent to the nonlinear Langevin equation, i.e., ϕ˙ = f (ϕ) + g(ϕ)L(t),

(8.97)

where f (ϕ) = −

dV (ϕ) = μϕ − λϕ3 . dϕ

The ground state of any system is determined by fluctuations whose mean values can be zero while the correlations are conserved. Moreover, an arbitrary system can be in contact with a nonlinear environment whose behavior is not fully unambiguous. A random migration of the system over various states is the result of both the direct action of the vacuum and random action due to the contact of the system with the nonlinear vacuum where it is located. A random influence of the vacuum can be taken into account only in the form of correlations between fluctuations at different time

February 5, 2021

300

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 300

Statistical Physics of Nonequilibrium Systems

instants L(t)L(t ) = φ(t − t ) because the mean values of such fluctuations are equal to zero. The mean value . . . of correlations is nonzero only during the time interval of the action. Therefore, the function φ(t − t ) should have a sharp peak as the time interval tends to zero, which corresponds to the condition φ(τ )dτ = σ 2 that is peculiar to white noise [242]. For this reason, it is more expedient to consider the definition of the nonequilibrium distribution functions of states from the nonlinear Fokker–Planck equation. Such approach is suitable for the description of the behavior of a nonequilibrium system where there occurs relevant energy income from outside, energy losses under the direct action of the environment, and energy dissipation under the random influence of the environment. Considering probable processes that may occur in the system itself and on exchange with the environment, we have to employ a more general approach with nonlinear Langevin equation. Though the form of the Langevin equation differs from that of the Fokker–Planck equation, they are mathematically equivalent [242]. The equation for the distribution function provides more information than the dynamic equation and can describe the probable phase transition to the new state. Assuming that the coefficient g(ϕ)) depends on the initial time we have to use the Fokker–Planck equation in the Ito form. If this coefficient depends on the control parameter (in our case, the fundamental scalar field) before and after the transition, then we have to apply the Fokker–Planck equation in the Stratonovich form [242], i.e., ∂ σ2 ∂ ∂ ∂ρ =− (f (ϕ)ρ) + g(ϕ) g(ϕ)ρ. ∂t ∂ϕ 2 ∂ϕ ∂ϕ

(8.98)

In what follows, we only use the Stratonovich representation, the more so, as both approaches are directly related [76, 242]. Both equations have no particular physical meaning until the physical process under consideration is specified. The above equation for the nonequilibrium distribution function of the system may be rewritten by conservation law for the distribution function that is

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 301

Statistical Description of Nonequilibrium Systems

301

given by ∂J(ρ(ϕ, t)) ∂ρ(ϕ, t) = , ∂t ∂ϕ where the probability flow may be written in the form

 σ2 ∂ σ2 ∂ J = − f (ϕ) − g(ϕ) g(ϕ) ρ + g2 (ϕ) ρ. 2 ∂ϕ 2 ∂ϕ

(8.99)

(8.100)

The general stationary solution of the Fokker–Planck equation in the absence of the flow is given by   ϕ 2f (ϕ )dϕ − ln g(ϕ) . (8.101) ρs (ϕ) = A exp σ 2 g2 (ϕ ) 0 The equilibrium distribution function given by the stationary solution associated with the nonlinear properties of the environment may be rewritten in the form ρs (ϕ) = A exp {−U (ϕ)} , where we have introduced the potential  ϕ 2f (ϕ )dϕ . U (ϕ) = ln g(ϕ) − σ 2 g2 (ϕ ) 0

(8.102)

(8.103)

This distribution function has the form that is not limited by the Gaussian distribution. If the dissipation in the system is described by a nonlinear function f (ϕ) of the scalar field, and the diffusion coefficient depends on the scalar field, a number of situations characterized by new equilibrium states of the nonequilibrium system can be realized. One of the probable ways when considering various fluctuations of the vacuum is to write μ2 = μ2 + ξ2 where the second term is associated with the random influence of the environment on the given coefficient. The dependence of the diffusion coefficient on the fundamental scalar may be obtained from the linear Langevin equation and using the theory of Markov processes making allowance for the nonequilibrium fluctuations of arbitrary coefficients in the function describing the direct action of the environment f (ϕ). In

February 5, 2021

302

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

page 302

Statistical Physics of Nonequilibrium Systems

our case, the second term is associated with the nonlinearity in the system. The standard approach may be given, as before, in terms of the Fokker–Planck equation as [242]:  σ2 ∂ 2 2 ∂  2 ∂ρ(ϕ, t) = μ ϕ − λϕ3 )ρ(ϕ, t) + ϕ ρ(v, t). ∂t ∂ϕ 2 ∂ϕ2 The stationary solution of the equation is given by   2λϕ2 (2μ2 /σ2 )−1 exp − 2 , ρs (ϕ) = N ϕ σ

(8.104)

(8.105)

which corresponds to the non-Gaussian distribution over the scalar field. The extremum of the distribution function now depends on the value of noise. If 0 < μ2 < σ 2 /2, then the stationary distribution function keeps the behavior of the delta function for ϕ = 0. When the parameter becomes greater under the dispersion of the multiplicative 2 > σ 2 /2, the distribution function attains maximum for ϕ = noise μ μ2 /λ and the Universe is associated with the zero scalar ϕ0 = parameter rather than nonzero Hubble parameter. The extremum of the distribution function is in agreement in this case with the zero of the new equation (μ2 −(σ 2 /2))ϕ−λϕ3 = 0, from which the conclusion follows that the model has two other points of transitions to the new states Fig. 8.2. The distribution function in the case of simple white noise g(ϕ) = 1 from Eq. (1.76) may be presented as   2V (ϕ) ρs (ϕ, t) = N exp − 2σ 2 that has an extremum only for ϕ = ϕ0 . If we take into accountprobable dependence of the Hubble parameter as a function, H = a V (ϕ), then we can solve the Fokker–Planck equation and find the stationary distribution function in the form   2H 2 ρs (ϕ, t) = N exp − 2 2 2a σ that has a sharp peak for ϕ = ϕ0 too. This result suggests a conclusion that the current inflation can be explained only by the

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Description of Nonequilibrium Systems

page 303

303

ρs (φ ) 6 5 4 3 2 1

0.0

Fig. 8.2.

0.5

1.0

1.5

2.0

λφ 2 μ2

Dependence of the distribution function on the noise level.

interaction with the multiplicative noise. The state with ϕ = 0 can occur here only in the above-mentioned case after observing the necessary condition for the formation of the new phase bubble. We have shown earlier that when all probable multiplicative fluctuations are taken into account and the probability of a transition into a stable vacuum state is calculated, various situations can be realized. For this reason, we may employ the interaction with the multiplicative noise and regard the probable changes and dispersion of fluctuations as being caused by the scalar field potential. In this case, the nonlinearity should be taken into account both in the potential as in the behavior of fluctuations of other nature. Probable fluctuations change the minimum of the potential and determine an alternative way for dynamical Universe formation. With this behavior, the state of the Universe in the present case can determine ϕ = 0 rather than nonzero Hubble constant. Thus, a model is proposed for describing the nonequilibrium Universe by the formation of new stationary states. The stationary distribution functions of the Universe are obtained for the standard mechanism of formation, in contact with the nonlinear environment. This statement

February 5, 2021

304

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-ch08

Statistical Physics of Nonequilibrium Systems

is well founded due to the presence of the negative mass coefficient in the fundamental scalar field potential. The negative mass coefficient gives rise to an instability of the ground state and favors the appearance of probable fluctuations that can be taken into account only in this approach.

page 304

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-con

Conclusions

The results presented in this book illustrate the possibility to describe the systems of interacting particles by their spatially inhomogeneous distributions using statistical theory. The representation of the partition function in terms of the functional integral over the auxiliary fields corresponds to the construction of an equilibrium sequence of probable states with respect to their weights. With the partition function being treated in this way, we may employ the methods of quantum field theory. The extension to the complex plane provides a possibility to apply the saddle-point method and thus to select the states whose contributions in the partition function are dominant. The solutions associated with the finite values of the “effective thermodynamic potential” functional may be regarded as thermodynamically stable particle distributions. Whether the distribution is homogeneous or inhomogeneous depends on the solutions that satisfy the extremum condition for the functional. Thus the spatially inhomogeneous distribution of the auxiliary fields may be unambiguously related to the spatially inhomogeneous particle distribution. It is also possible to find the parameters of such formations and the temperature of the phase transition accompanied by the formation of finite-size clusters of the new phase. Actually, this approach extends the mean-field approximation to consider spatially inhomogeneous field distributions. The approach proposed here is an advancement in the study of the behavior of clusters formed. In the case of multisoliton solutions, the residual interaction (uncompensated in the course

305

page 305

February 5, 2021

306

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-con

Statistical Physics of Nonequilibrium Systems

of cluster formation) produces new spatial structures. The soliton interaction energy is described by an expression of the form ωrr = A exp[−k(r − r  )] [33, 65]. Obviously, this system of clusters may be regarded as a gas of interacting particles and traditional methods of statistical physics, e.g., the simplest Ising model, may be employed to estimate the temperature of the phase transition to the spatially ordered state. It is also possible then to consider the collective behavior of such formations. In the approach considered here, there is no need to introduce two auxiliary fields that correspond to the attraction and repulsion, respectively. We may introduce one complex field ϕ + iψ associated with the interaction of any type, and carry out the procedure in the complex plane. We only have to know the inverse operator of the interaction. Dividing the interaction into several parts provides a better understanding of the mechanisms of spatially inhomogeneous particle distribution formation. Actually, this method describes the first-order phase transitions to the states that contain new phase bubbles. In this book, based on the new method, we have studied the properties of the model system of a self-gravitating gas with fixed number of particles N and energy. The size of the cluster is determined by the balance of gravitation and thermal energies, so that with the increase of the temperature, the size of the cluster increases while the density in the center decreases. The size is determined by the minimum of the free energy, however, the relevant minimum is not absolute and the process of collapse may continue further while each cluster acts as a particle. This situation is analogous to the false vacuum in field theory [24]. In the case of large but finite size of the system and the number of particles in it, a value of temperature exists such that if the temperature of the system is higher than this value, then the gas cannot collapse. Moreover, such a value of concentration exists so that if the concentration of the system is lower than this value, then the gas cannot collapse too. The collapse cannot occur for any degree of compression if the temperature is higher than the critical temperature. Our results show that the gravitation interaction of particles leads to the formation of a cluster of finite size, as the initial

page 306

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-con

Conclusions

page 307

307

homogeneous state is unstable in the limiting case N → ∞, V → ∞, but N/V is fixed. The equation for the spatial distribution function is obtained, and the size of the spatial inhomogeneity is determined by the thermodynamic conditions. The characteristic sizes of the two media, i.e., the gravitating gas and repulsing particles (by Coulomb) are similar to the same interaction constant in the Boltzmann limit and the degenerated case. The reason for this coincidence is that the spatial distributions in both cases are determined by the two energies only, i.e., the Coulomb-type and thermal energy or the repulsive Fermi “statistical potential”. The mechanisms of clusters and polarons formation are similar — the balance of the two energies. This fact may be explained in terms of the hydrodynamic approach — both media have equal time scales of the structure formation, but in the first case, it determines the relaxation time while in the second case, it determines the plasma frequency. These time scales determine spatial scales (the radii of clusters and polarons) monotonously. To solve the problem of the divergence of the partition function for a gravitating system is impossible. We have shown, however, that this difficulty may be avoided and hence the state with multitude of clusters may be treated as metastable. The gravitation interaction of particles results in the cluster formation of finite size, as the initial homogeneous state is unstable. The size is determined by the thermodynamic conditions, in particular, the increase of temperature is accompanied by the increase of the cluster size that causes the decrease of the mean density; the cluster size decreases when particles are added in the system that is associated with the increase of the gravitation energy. Such behavior is caused by the long-range attraction of the gravitation interaction (1/R). The size of the cluster approaches the equilibrium size asymptotically in the course of its formation. The state of a system with a spatially inhomogeneous distribution function corresponding to the cluster of equilibrium size is stable. Unfortunately, the results of the study of this model cannot be verified experimentally. Nevertheless these results (cluster formations within the Boltzmann limit) may be useful in problems of

February 5, 2021

308

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-con

Statistical Physics of Nonequilibrium Systems

astrophysics, for the study of the formation of giant planets, stars, and the accumulation of the gas-dust matter, in particular. We have developed a formalism to describe the spatially inhomogeneous distribution in a system of interacting particles. It employs the new unconventional method that uses the Hubbard–Stratonovich representation of the partition function. The method is extended and applied to a system with Coulomb-like interaction to find the solution for the particle distribution. It is important that this solution has no divergences for the thermodynamic limiting cases. We use the saddlepoint approximation by considering the conservation of the number of particles that yields a nonlinear equation for the new field variable. In the three-dimensional case, this equation reduces to the sineGordon equation whose solution determines the state associated with the dominant contribution in the partition function. This approach helps to describe the conditions of Wigner crystal formation in a system of dust particles in a plasma. Various possibilities may be probable for different parameters corresponding to the interaction potential. However, the results for simple and basic cases are very important in the understanding of the behavior of a dusty plasma in situations involved. We have analytically derived the necessary condition for the crystal formation in a system of dust particles in the three-dimensional case. In the one- and two-dimensional cases, we have found exact solutions for various spatial distributions of charged particles. The proposed method is designed for the studies of selfassembling systems where nonuniform distributions are found for different scale lengths. We present the partition function and the equation of state in the most general form. In the threedimensional case we obtained a condition (quadratic dependence of the coupling parameter on the experimental parameter l) that is in good agreement with the experimental data reported in [34]. In the one-dimensional case, we obtained a structure of the periodic distribution of charged particles along a cylindrical sample. In the two-dimensional case, an exact result is obtained for the partition function of the homogeneous distribution in a purely Coulomb system.

page 308

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-con

Conclusions

page 309

309

The approach proposed in this book makes it possible to describe the first-order phase transitions accompanied by the formation of new-phase macrobubbles at both microscopic and phenomenological levels. The selection of states that bring dominant contributions in establishing the stable thermodynamic behavior, making allowance for fluctuations at both levels, provide a complete picture of processes that accompany the first-order phase transitions. The state selection is combined with the calculation of the saddle-point configuration of the fields that contribute the most towards the thermodynamic characteristics whereas the fluctuations against the background of the order parameter treated as the slowest subsystem enables us to consider fluctuations into account for both cases of small-scale and long-wave variations. The separation of scales occurs under the formation of a finite-size nucleus of the new phase. The fluctuations of the scale smaller than the bubble size may be taken into account by averaging over the fast variable. Thus the treatment is reduced to considering a new effective potential regarding the collective condensate behavior of fast fluctuations. The large-scale fluctuations whose variation scale is larger than the spatial rate of change of the order parameter act as an external field with respect to the bubbles formed, so the dispersion of changes thereof determines the average size of the new phase bubble. Thus it is possible to consider fluctuations of any scale by a unified approach given that fast fluctuations produce the nonlinear potential for the order parameter that just serves to form the new phase bubble, while slower fluctuations determine the size of the latter that can be observed experimentally. Separation of fluctuations of a certain type is governed mainly by the behavioral specifics of the system at the microscopic level; separation of the main governing parameters is associated in the long run with the character and value of the interparticle interaction in the system. The correct treatment of the microscopic behavior of the system determines all the governing parameters of the system along with the characteristics of the firstorder phase transition accompanied by the formation of new-phase macrobubbles.

February 5, 2021

310

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-con

Statistical Physics of Nonequilibrium Systems

The equilibrium distribution of interacting particles corresponds to their spatially nonuniform distribution. This spatial nonuniform distribution of the introduced particles creates areas free of particles. The combined effect of particles and the medium produces a nonuniform distribution of cooperating particles. Some kind of a new soft body is formed whose properties differ from the properties of the medium. It is possible to form a wide variety of cellular structures if the forces of interaction possess anisotropy. Another important property of such systems is that they are highly sensitive to weak external effects. This enables us to exert a specific effect upon the formation of structures and their transmutations. Moreover, this system has got an advantage of being visually observable because one can directly observe all the changes in the structures. Thus, a general description of the formation of a cellular structure in the system of interacting particles is proposed. The analytic results have been presented for probable cellular structures in ordinary colloidal systems, in systems of particles immersed in a liquid crystal, and in gravitation systems. The formation of cellular structures in all systems of interacting particles for different temperatures and concentrations of particles is shown to have the same physical nature and have the same theoretical description. Interacting particle systems can be nonequilibrium a priori. Before relaxing towards thermodynamic equilibrium, isolated systems with long-range interactions are trapped in nonequilibrium quasi-stationary states whose lifetimes diverge as the number of particles increases. A theory is presented that makes it possible to quantitatively predict the instability threshold for spontaneous symmetry breaking for a class of d-dimensional systems. Nonequilibrium stationary states of three-dimensional systems do not evolve to thermodynamic equilibrium but are trapped in nonequilibrium quasi-stationary states. The presented nonequilibrium statistical description deals only with the probable structures that occur in a self-gravitating system. It does not describe the states and tells nothing about time scales of the kinetic theory. The statistical operator has no singularities for any value of the gravitation field. The challenge to describe a self-gravitating particle system can be solved

page 310

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-con

Conclusions

page 311

311

by this approach provided we take into account inhomogeneous particle and temperature distributions. The gravity factor may either favor or retard such transformations; this depends on the character of the system under consideration and the relevant conditions. For the first time, we have described the formation of spatially inhomogeneous particle distributions followed by the changes of the temperatures of these distributions of interacting particles. The statistical description of the system was tailored to the particles with gravitational interaction by an arbitrary spatially inhomogeneous particle distribution. In this approach, the behavior of a selfgravitating system can be predicted for any external conditions. In this way, one can develop a statistical description of a self-gravitating system. We propose an approach that provides a possibility to quantitatively predict the particle distribution in a system with special repulsive interaction. In this way, we can solve the complicated problem of the statistical description of systems with special repulsive interactions and introduce a new field variable that reduces this task to the solution of the cosmological problem. Moreover, this method may also be applied for the further development of physics of selfgravitating and related systems that are not far from equilibrium. Another important conclusion is that we can calculate the partition function by integration over the energy. Such integration in this sense implies a continual integral over the energy variable. The extremum of the partition function is realized under the thermodynamic condition and any probable deviation from this condition contributes very little to the macroscopic characteristics similar to the quantum contribution to classical trajectories. The novelty of this presentation consists in the possibility to use the kinetic equations for the description of nonequilibrium systems such as Brownian systems in the energy space. Starting from the basic kinetic equations for the distribution function of the macroscopic system in the energy space, we can obtain steady states and fluctuation dissipation relations for nonequilibrium systems. A probable approach to the description of nonequilibrium states has been proposed too. Based on the Fokker– Planck equation for the nonequilibrium distribution functions of macroscopic systems, a stationary solution was obtained that may be

February 5, 2021

312

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-con

Statistical Physics of Nonequilibrium Systems

interpreted as the “equilibrium” distribution function for the new energy state. Such approach takes into account probable motion between different states of the system, induced by the dissipation of energy and influence of the environment that depends on the energy of the system. A nonlinear model is proposed representing probable stationary states of systems with various spatial processes within them. Acknowledgments This book was prepared with the financial support of the National Academy of Sciences of Ukraine within the research grants “The structure and dynamics of statistical and quantum field systems,” “Noise-inducing dynamics and correlations in nonequilibrium systems” and “Mathematical models of non-equilibrium processes in the open systems”. The authors are also very grateful to Dr. Olga Kocherga for translating and editing the book into English and to Ms. Zoya Vakhnenko for the help in preparing the manuscript.

page 312

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-bib

Bibliography

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24]

E. D. Belotskii, B. I. Lev. Teor. Mat. Fiz. (Sov) 60, 121 (1984). E. D. Belotskii and B. I. Lev. Phys. Lett. 147, 13 (1990). B. I. Lev and A. Ya. Zhugaevych. Phys. Rev. E 57, 6460 (1998). B. I. Lev and A. G. Zagorodny. Phys. Lett. A 373, 1101–1104 (2009). K. V. Grigorishin and B. I. Lev. Phys. Rev. E 71, 066105 (2005). B. I. Lev and K. V. Grigorishin. Cond. Matt. Phys 45, 1 (2006). H. Kleinert. Gauge Fields in Condensed Matter. World Scientific, Singapore, 1989. H. J. de Vega, N. Sanches, and F. Combes. Phys. Rev. D 54, 6008 (1996). L. N. Lipatov. JETP (Sov) 72, 412 (1977). S. Edward and A. Lenard, J. Math. Phys. 3, 778 (1962). J. Hubbard. Phys. Rev. Lett. 3, 77 (1959). R. L. Stratonovich. Sov. Phys. Dokl. 2, 416 (1984). S. Samuel. Phys. Rev. D 18, 1916 (1978). V. B. Magalinsky. JETP (Sov) 48, 167 (1965). H. Kleinert, Fortschr. Phys. 26, 565 (1979). K. Rose, E. Gurewitz, and G. C. Fox. Phys. Rev. Lett. 65, 945 (1990). H. M. Ellerby. Phys. Rev. E 49, 4287 (1994). K. S. Min, K. V. Shcheglov, C. M. Yang, and H. A. Atwater. Phys. Rev. Lett. 69, 2033 (1996). B. S. Edelman. Usp. Fiz. Nauk (Sov) 130, 675 (1980). B. J. Melnikov and C. V. Meschkov, JETP Lett. 33, 222 (1981). Jo. E. Gegusin, Ju. S. Kaganovskii, and B. B. Kalinin. Phys. Stat. Solid. 11, 250 (1969). W. Skorupa, R. A. Yankov, and I. E. Tyschenko. Appl. Phys. Lett. 68(17), 2410 (1996). E. D. Belotskii. Doctor Degree Thesis, Institute of Physics, Kiev, 1992. S. Coleman. Phys. Rev. D 15, 2929 (1977). 313

page 313

February 5, 2021

314

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-bib

Statistical Physics of Nonequilibrium Systems

[25] S. Ramaswamy, R. Nityananda, V. A. Raghunathan, and J. Prost. Molecular Crystals and Liquid Crystals 288, 175 (1996). [26] S. Edward and A. Lenard. J. Math. Phys. 3, 778 (1962). [27] C. C. Bradley, C. A. Sackett, J. J. Tollett, and R. G. Hulet. Phys. Rev. Lett. 75, 1687 (1995). [28] K. B. Davis, M. O. Mewes, M. R. Andrews, N. J. van Druten, D. S. Durfee, D. M. Kurn, and W. Ketterle. Phys. Rev. Lett. 75, 3969 (1995). [29] C. C. Bradley, C. A. Sackett, and R. G. Hulet. Phys. Rev. Lett. 78, 985 (1997). [30] L. P. Pytaevsky. Zh. Eksp. Teor. Fiz. 40, 646 (1961). [31] E. P. Gross. J. Math. Phys. 4, 195 (1963). [32] P. A. Ruprecht, M. J. Holland, and K. Burnett. Phys. Rev. A 51, 4704 (1995). [33] R. Rajaraman. Phys. Rev. 15, 2866 (1977). [34] V. E. Fortov, A. V. Ivlev, S. A. Khrapak, A. G. Khrapak, and G. E. Morfill. Phys. Rep. 421, 1 (2005). [35] H. Lowen. Phys. Rep. 237, 249 (1994). [36] G. E. Morfill, H. M. Thomas, U. Konopka, and M. Zuzic. Phys. Plasmas 6, 1769 (1999). [37] E. Helpand and F. H. Stilinger. J. Chem. Phys. 49, 1232 (1968). [38] B. I. Lev, V. B. Tymchyshyn, and A. G. Zagorodny. Condens. Phys. 12, 593 (2009). [39] H. Thomas, G. E. Morfill, V. Demmel, J. Goree, B. Feuerbacher, and D. Mohlmann. Phys. Rev. Lett. 73, 652 (1994). [40] J. H. Chu and I. Lin. Phys. Rev. Lett. 72, 4009 (1994). [41] A. Melzer, T. Trottenberg, and A. Piel. Phys. Lett. A 191, 301 (1994). [42] A. G. Sitenko, A. G. Zagorodny, and V. N. Tsytovich. AIP Conf. Proc. 345, 311 (1995). [43] S. A. Brazovsky. Sov. Phys. JETP 68, 715 (1975). [44] B. I. Lev and H. Yokoyama. Int. J. Mod. Phys. B 17, 4913 (2003). [45] H. Totsuji, T. Kishimoto, and C. Totsuji. Phys. Rev. Lett. 78, 3113 (1997). [46] S. V. Vladimirov, S. A. Khrapak, M. Chaudhuri, and G. E. Morfill. Phys. Rev. Lett. 100, 055002 (2008). [47] H. Ikezi, Phys. Fluids. 29, 1764 (1986). [48] A. Melzer, A. Homann, and A. Piel. Phys. Rev. E 53, 2757 (1996). [49] J. C. Crocker and D. G. Grier, Phys. Rev. Lett. 77, 1897 (1996). [50] R. Aveyard, B. P. Binks, J. H. Clint, P. D. I. Fletcher, T. S. Horozov, B. Neumann, V. N. Paunov, J. Annesley, S. W. Botchway, D. Nees,

page 314

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-bib

Bibliography

[51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70]

[71] [72] [73] [74] [75]

page 315

315

A. W. Parker, A. D. Ward, and A. N. Burgess. Phys. Rev. Lett. 88, 246102 (2002). R. R. Netz. Phys. Rev. E 60, 3174 (1999). D. Ruelle. Statistical Mechanics: Rigorous Results. Benjamin, New York, 1969. K. Huang. Statistical Mechanics. J. Wiley and Sons, New York, 1963. A. Isihara. Statistical Mechanics. State University of New York, 1971. J. Ortner. Phys. Rev. E 59, 6312 (1999). A. Ciach, Phys. Rev. E 78, 061505 (2008). B. I. Halperin and T. C. Lubensky. Phys. Rev. Lett. 32, 292 (1974). Yu. Kagan, G. V. Shlyapnikov, and J. T. M. Walraven. Phys. Rev. Lett. 76, 2670 (1996). B. I. Lev and A. G. Zagorodny. Eur. Phys. J. B 86, 422 (2013). G. B. Whitham. Linear and Nonlinear Waves. J. Wiley and Sons, New York, 1974. G. Stell, K. C. Wu, and B. Larsen. Phys. Rev. Lett. 37, 1369 (1976). B. I. Lev, V. B. Tymchyshyn, V. K. Ostroukh, and A. G. Zagorodny. Eur. Phys. J. B 87, 253 (2014). R. Tao, J. T. Woestman, and N. K. Jaggi. Appl. Phys. Lett. 55, 1844 (1989). R. Tao and J. M. Sun. Phys. Rev. Lett. 67, 398 (1991). L. D. Landau and E. M. Lifshitz. Statistical Physics. Pergamon Press, Oxford, 1959. J. S. Langer. Ann. Phys. 41, 108 (1967). D. C. Hobenberg, B. I. Halperin. Rev. Mod. Phys. 49, 435 (1977). A. G. Khachaturyan. Theory of Phase Transformation and the Structure of the Solid Solutions. Nauka, Moscow, 1974 (in Russian). J. M. Ziman. Models of Disorder. Cambridge University Press, Cambridge, 1979. A. Z. Patashinskii and B. L. Pokrovskii. Fluctuation Theory of Phase Transitions. Nauka, Moscow, 1975, 1982 (in Russian). Pergamon Press, Oxford, 1979. P. M. Platzman and H. Fukuyama. Phys. Rev. B 10, 3150 (1974). F. M. Peeters and P. M. Platzman. Phys. Rev. Lett. 50, 2021 (1983). E. M. Lifshitz and L. P. Pitajevskii. Physical Kinetics. Nauka, Moscow, 1979 in Russian); Pergamon Press, Oxford, 1981. A. F. Chaykin and T. C. Lubensky. Principles of Condensed Matter Physics. Cambridge University Press, Cambridge, 1995. A. C. Zefflemoyer (Ed). Nucleation. Dekker, New York, 1969.

February 5, 2021

316

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-bib

Statistical Physics of Nonequilibrium Systems

[76] W. Horsthemke and R. Lefever. Noise-Induced Transition: Theory, Applications in Physics, Chemistry and Biology. Springer-Verlag, New York, 1984. [77] A. H. Guth and E. Weinberg. Nucl. Phys. B 212, 321 (1983). [78] A. D. Linde. Particle Physics and Inflationary Cosmology. Chur: Switzerland, Harwood Academic Publishers, 1990. [79] G. E. Volovik and U. E. Dzjaloschinskii. JETP 75, 1102 (1978). [80] P.-H. Chavanis. Int. J. Mod. Phys. B 20, 3113 (2006). [81] W. C. Saslow. Gravitational Physics of Stellar and Galactic System. Cambridge University Press, New York, 1987. [82] P. Tarazona. Phys. Rev. A 31, 2672 (1985); P. Tarazona. Phys. Rev. Lett. 84, 694 (2000); P. Tarazona and Y. Rosenfeld. Phys. Rev. E 55, R4873 (1997). [83] H. Lowen. Phys. Rev. Lett. 74, 1028 (1995). [84] S. H. Chen and N. M. Amer. Phys. Rev. Lett. 51, 9298 (1983). [85] B. J. Liang and S. C. Chen. Phys. Rev. A 39, 1441 (1989). [86] D. A. Soville, W. B. Russel and W. R. Schowaiter. Colloidal Dispersions. Cambridge University Press, Cambridge, 1989. [87] T. Padmanabhan. Phys. Reports 188, 285 (1990). [88] S. Chandrasekhar. Principle of Stellar Dynamics. Dover Publication, New York, 1960. [89] W. Thirring. Z. Phys. 235, 339 (1970). [90] C. Sire and P.-H. Chavanis. Phys. Rev. E 66, 046133 (2002). [91] P.-H. Chavanis. Phys. Rev. E 65, 056123 (2002). [92] H. J. de Vega and N. Sanchez. Phys. Lett. B 490, 180 (2000); H. J. de Vega and N. Sanchez. Nucl. Phys. B 625, 409 (2002); H. J. de Vega and N. Sanchez. Nucl. Phys. B 625, 460 (2002). [93] V. Laliena. Nucl. Phys. B 668, 403 (2003). [94] R. Pakter, B. Marcos, and Y. Levin. Phys. Rev. Lett. 111, 230603 (2013). [95] F. P. C. Benetti, A. C. Ribeiro-Teixeira, R. Pakter, and Y. Levin. Phys. Rev. Lett. 113, 100602 (2014). [96] D. Zubaryev. Nonequilibrium Statistical Thermodynamics. Nauka, Moscow, 1971. [97] A. R. Denton, N. W. Ashcroft, and W. A. Curtin. Phys. Rev. E 51, 65 (1995). [98] Y. Rosenfeld, M. Schmidt, H. Lowen, and P. Tarazona. Phys. Rev. E 55, 4245 (1997); Y. Rosenfeld. Phys. Rev. Lett. 63, 980 (1989). [99] C. Rascon, L. Mederos, and G. Navascues. Phys. Rev. E 53, 5698 (1996).

page 316

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-bib

Bibliography

page 317

317

[100] B. B. Laird and D. M. Kroll. Phys. Rev. A 42, 4810 (1980). [101] S. C. Kim. J. Chem. Phys. 106, 1146 (1997). [102] W. K. Keget, H. Retas, and H. N. W. Lekkerkerker. Phys. Rev. Lett. 83, 5298 (1999). [103] I. Ispolatov and E. G. D. Cohen. Phys. Rev. Lett. 87, 210601 (2001). [104] B. C. Eu and K. Rah. Phys. Rev. E 63, 031203 (2001). [105] X. Z. Wang. Phys. Rev. E 66, 031203 (2002). [106] P. Richard, L. Oger, J. P. Troadee, and A. Gervois. Phys. Rev E 60, 4551 (1999). [107] C. Caccamo, G. Pellicane, D. Costa, D. Pini, and G. Stell. Phys. Rev. E 60, 5533 (1999). [108] G. Rybicki. Astrophys. Space Sci. 14, 56 (1971). [109] K. R. Yawn and B. N. Miller. Phys. Rev. E 68, 056120 (2003). [110] J. J. Aly. Phys. Rev. E 39, 3771 (1994). [111] B. I. Lev, S. S. Rozhkov, and A. G. Zagorodny. EPL 111, 26003 (2015). [112] Y. Levin, R. Pakter, F. B. Rizzato, T. N. Teles, and F. P. C. Benetti. Phys. Reports 535, 1 (2014). [113] M. Joyce and T. Worrakitpoonpon. J. Stat. Mech. 2010, P10012 (2010). [114] R. Michie. Mon. Not. R. Astron. Soc. 125, 127M (1963). [115] R. Michie and P. H. Bodenheimer. Mon. Not. R. Astron. Soc. 126, 269M (1963). [116] W. Jaffe. Mon. Not. R. Astron. Soc. 202, 995J (1983). [117] R. Balesku. Equilibrium and Nonequilibrium Statistical Mechanics. J. Wiley and Sons, New York, 1978. [118] J. A. S. Lima, R. Silva, and J. Santos. Astron. Astrophys. 396, 309 (2002). [119] A. P. Boss. Science 276, 1836 (1997). [120] S. M. Fall. Annals of the New York Academy of Science 336, 172 (2006). [121] B. I. Lev. Int. J. Mod. Phys. B 25, 2237 (2011). [122] D. C. Wang and A. P. Gast. Phys. Rev. E 59, 3964 (1999); C. Rascon, L. Mederos and G. Navascues. Phys. Rev. E 53, 5698 (1996). [123] S. Hess and M. Kroger. Phys. Rev. E 64, 011201 (2001). [124] U. F. Edgal and D. L. Huber. J. Chem. Phys. 108, 1578 (1998). [125] C. N. Likos, A. Lang, M. Watzlawek, and H. Lowen. Phys. Rev. E 63, 031206 (2001). [126] N. M. Dixit and C. F. Zukoski. Phys. Rev. E 64, 041604 (2001). [127] J. Bergenholtz and M. Fuchs. Phys. Rev. E 59, 5706 (1999).

February 5, 2021

318

[128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] [139] [140] [141] [142] [143] [144] [145] [146] [147] [148] [149] [150] [151] [152] [153] [154] [155] [156] [157] [158] [159]

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-bib

Statistical Physics of Nonequilibrium Systems

B. Groh. Phys. Rev. E 61, 5218 (2000). V. Kobelev and B. Kolomeisku. J. Chem. Phys. 116, 7589 (2002). J. A. Anta and S. Lago. J. Chem. Phys. 116, 10514 (2002). J. H. Herrera, H. Ruiz-Estrada, and L. Blum. J. Chem. Phys. 104, 6327 (1996). V. Babin, A. Ciach, and M. Tasinkevych. J. Chem. Phys. 114, 9685 (2001). J. M. Kosterliz and D. J. Thouless. J. Phys. C 6, 1181 (1973); K. J. Stranburg. Rev. Mod. Phys. 60, 161 (1988). P. G. de Gennes and J. Prost. The Physics of Liquid Cristals. Clarendon Press, Oxford (ed. 2), 1993. F. Brochard and P. G. De Gennes. J. Phys. (Paris) 31, 691 (1970). P. Aviles and Y. Giga. Proc. Center Math. Anal. Austr. Nat. Univer. 12, 1 (1987). P. Aviles and Y. Giga. Proc. Roy. Soc. Edin. Sect. A 129, 1 (1999). G. Gioia and M. Ortiz. Adv. Appl. Mech. 33, 119 (1997). M. Ortiz and G. Gioia. J. Mech. Phys. Solids 42, 531 (1994). N. M. Ercolani, R. Indik, A. C. Newell, and T. Passot. J. Nonlin. Sci. 10, 223 (2000). R. A. Cowley. Phys. Rev. B 13, 4877 (1985). M. Cross and A. Newell. Physica D 10, 299 (1984). A. Newell, T. Passot, C. Bowman, N. Ercolani, and R. Indik. Physica D 97, 185 (1996). A. E. Lobkovsky and T. A. Witten. Phys. Rev. E 55, 1577 (1997). S. B. Goryachev. Phys. Rev. Lett. 72, 1815 (1994). B. H. Liu and L. McLerran. Phys. Rev. D. 46, 2668 (1992). Y. Rubin and M. Gifferman. Phys. Rev. A. 29, 1496 (1984). R. Feynman. Statistical Mechanics. W.A. Benjamin, California, 1972. R. J. Baxter. Exactly Solved Models in Statistical Mechanics. Academic Press, London, New York, Sydney, Tokyo, Toronto, 1982. S. F. Edwards. Phil. Mag. 4, 1171 (1959). Y. S. Wu. Phys. Rev. Lett. 73, 922 (1994). F. D. M. Haldane. Phys. Rev. Lett. 67, 937 (1991). V. V. Krasnogolovetz and B. I. Lev. Ukr. Fiz. J. 39, 296 (1994). B. I. Lev. Phys. Rev. E 58, R2681 (1998). S. Doniach. Phys. Rev. B 24, 5063 (1981). P. Kampf and T. Zimanyi. Phys. Rev. B 47, 279 (1993). J. Jeans. Astronomy and Cosmogony. Cambridge Univ. Press, 1928. B. I. Lev. Eur. Phys. J. Special Topics 216, 37–48 (2013). P. Arnold. Phys. Rev. D. 46, 2628 (1992).

page 318

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-bib

Bibliography

[160] [161] [162] [163] [164] [165] [166] [167] [168] [169] [170] [171] [172] [173] [174] [175] [176] [177] [178] [179] [180] [181] [182] [183] [184] [185] [186]

page 319

319

A. D. Linde. Pep. Prog. Phys. 42, 389 (1979). A. P. Rebesh and B. I. Lev. Phys. Lett. A 218, 1–6 (2017). V. I. Klyatskin and D. Gurarie. Usp. Fiz. Nauk 169, 171 (1999). B. B. Kadomtzev and M. B. Kadomtzev. Usp. Fiz. Nauk 167, 649 (1997). Yu. Kagan, A. M. Muryshev, and G. V. Shlyapnikov. Phys. Rev. Lett. 81, 933 (1998). Yu. Kagan, E. L. Surcov, and G. V. Shlyapnikov. Phys. Rev. A 54, R1753 (1996). A. Gammal, L. Tomo, and T. Frederco. Phys. Rev. A 66, 043619 (2002). G. S. Jeon, L. Yn, S. W. Rhee, and D. J. Thouless. Phys. Rev. A 66, 011603(R) (2002). W. Ketterle. Usp. Fiz. Nauk 173, 1339 (2003). L. Parker and Y. Zhang, Phys. Rev. D 44, 2421 (1991). V. A. Antonov. Vestn. Leningr. Gos. Univ. 7, 135 (1962). D. Lynden-Bell and R. Wood. Mon. Not. R. Astron. Soc. 138, 495 (1968). M. Kiessling. J. Stat. Phys. 55, 203 (1989). B. Stahl, M. Kiessling, and K. Schindler. Planet. Space Sci. 43, 271 (1995). E. B. Aronson and C. J. Hansen. Astrophys. J. 177, 145 (1972). E. Follana and V. Laliena. Phys. Rev. E 61, 6270 (1999). V. Latora, A. Rapisarda, and S. Ruffo. Phys. Rev. Lett 80, 692 (1998). M. Antoni and A. Torcini. Phys. Rev. E 57, R6233 (1998). A. J. Vainschtein, V. I. Zakharov, and M. A. Shifman. Usp. Fiz. Nauk 136, 553 (1982). J. Ruiz-Garsia, R. Gamez-Corrawe, and B. I. Ivlev. Phys. Rev. E 58, 960 (1998). V. G. Anderson, E. M. Terentjev, S. P. Meeker, J. Crain, and W. G. K. Poon. Eur. Phys. E 4, 11 (2001). V. G. Anderson and E. M. Terentjev. Eur. Phys. E 4, 21(2001). J. Loudet, P. Barois, and P. Poulin. Nature 407, 611 (2000). P. Poulin, H. Stark, T. C. Lubensky, and D. A. Weitz. Science 205, 5770 (1997). P. Poulin and D. X. Weitz. Phys. Rev. E 57, 626 (1998). V. Nazarenko, A. Nych, and B. Lev. Phys. Rev. Lett 87, 075504 (2001). A. Nych, U. Ognysta, M. Skarabot, M. Ravnik, S. Zumer, and I. Musevic. Nature Communications 4, 1489 (2013).

February 5, 2021

320

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-bib

Statistical Physics of Nonequilibrium Systems

[187] R. W. Ruhwandl and E. M. Terentjev. Phys. Rev. E 55, 2958 (1997). [188] S. B. Chernyshuk, B. I. Lev, and H. Yokoyama. Zh. Eksp. Teor. Fiz. [Sov. Phys. JETP] 120, 431 (2001). [189] B. I. Lev. Ukr. J. Phys. Reviews 2(1), 3–34 (2005). [190] B. I. Lev, P. M. Tomchuk, and K. M. Aoki. Condens. Matter Phys. 6, 166 (2005). [191] B. I. Lev, S. B. Chernyshuk, P. M. Tomchuk, and H. Yokoyama. Phys. Rev. E 65, 021603 (2002). [192] S. B. Chernyshuk, B. I. Lev, and H. Yokoyama. Phys. Rev. E 71, 062701 (2005). [193] B. I. Lev, K. M. Aoki, and H. Yokoyama. Molecular Crystals and Liquid Crystals Science and Technology. Section A: Molecular Crystals and Liquid Crystals 367, 537–544 (2001). [194] F. Ghezzi and J. C. Earnshaw. J. Phys.: Condens. Matter 9, L517 (1997). [195] A. Borstnik, H. Stark, and S. Zumer. Phys. Rev. E 60, 4210 (1999). [196] P. Galatola and J. B. Fournier. Phys. Rev. Lett. 86, 3915 (2001). [197] J. B. Fournier and P. Galatola. Phys. Rev. E 65, 031601 (2002). [198] T. C. Lubensky, D. Pettey, N. Currier, and H. Stark. Phys. Rev. E 57, 610 (1998). [199] D. Pettey, T. C. Lubensky, and D. R. Link. Liquid Crystals 25, 579 (1998). [200] S. L. Lopatnikov and V. A. Namiot. Zh. Eksp. Teor. Fiz. 75, 361 (1978). [201] P. Poulin, V. Cabuil, and D. A Weitz. Phys. Rev. Lett. 79, 4862 (1997). [202] P. Poulin, V. A. Raghunathan, P. Richetti, and D. Roux. J. Phys. II France, 4(9), 1557 (1994). [203] S. V. Burylov and Yu. L. Raikher. J. Magn. Magn. Mater. 122, 62 (1993). [204] S. V. Burylov and Yu. L. Raikher. Phys. Rev. E 50, 358 (1994). [205] I. Janossy, A. D. Lloyd, and B. S. Wherrett. Mol. Cryst. Liq. Cryst. 179, 1 (1990). [206] D. Hilbert. Die Grundlagen der Physik. Nachrichten von der Gesellschaft der Wissenschaften zu G¨ ottingen — MathematischPhysikalische Klasse. Nachrichten, 395 p., 1915. [207] A. Einstein. Die Grundlage der Allgemeinen Relativitatstheorie. Annalen der Physik 354(7) 769–822 (1916). [208] S. Weinberg. Rev. Mod. Phys. 46, 255 (1974). [209] M. Ikeda and Y. Miyachi. Prog. Theor. Phys. 27, 474 (1962).

page 320

February 5, 2021

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-bib

Bibliography

page 321

321

[210] T. T. Wu and C. N. Yang. Properties of Matter under Unusual Conditions. (H. Mark and S. Fernbach, Eds.). Wiley-Interscience, New York, 1969. [211] R. Franzosi, M. Pettini, and L. Spinelli. Phys. Rev. Lett. 84, 2774 (2000). [212] L. Casetti and A. Macchi. Phys. Rev. E 55, 2539 (1997). [213] L. Caiani, L. Casetti, C. Clementi, G. Pettini, M. Pettini, and R. Gatto. Phys. Rev. E 57, 3886 (1998). [214] R. Franzosi, L. Casetti, L. Spinelli, and M. Penttini. Phys. Rev. E 60, R5009 (1999). [215] L. Casetti, R. Livi, and M. Pettini. Phys. Rev. Lett. 74, 375 (1995). [216] J. Jurkowski. Phys. Rev. E 62, 1790 (2000). [217] K. Kaviani, A. Dalafi-Rezaie. Phys. Rev. E 60, 3520 (1999). [218] G. Ruppeiner. Phys. Rev. A 44, 3583 (1991). [219] G. Ruppeiner. Phys. Rev. E 57, 5135 (1998). [220] V. Periwal. Phys. Rev. Lett. 78, 4671 (1997). [221] R. Blumenfeld. Phys. Rev. Lett. 78, 1203 (1997). [222] G. Garcia de Polavieja. Phys. Rev. Lett. 81, 1 (1998). [223] W. T. Gozdz and G. Gompper. Phys. Rev. E 59, 4305 (1999). [224] G. Foltin. Phys. Rev. E 57, R3742 (1998). [225] M. V. Jari´c (ed.). Aperiodicity and Order. Volume 1: Introduction to Quasicrystals. Academic Press, 1988. [226] M. Seul and D. Adelman. Science 267, 476 (1995). [227] B. I. Lev. Mod. Phys. Lett. B 27, 1330020 (2013). [228] E. L. Altschuler, T. J. Williams, E. R. Ratner, F. Dowla, and F. Wooten. Phys. Rev. Lett. 72, 2671 (1994). [229] E. L. Altschuler, T. J. Williams, E. R. Ratner, R. Tipton, R. Stong, F. Dowla, and F. Wooten. Phys. Rev. Lett. 78, 2681 (1997). [230] M. Bowick, A. Cacciuto, D. R. Nelson, and A. Travesset. Phys. Rev. Lett. 89, 185502 (2002). ´ [231] V. S. Edel’man. Levitated Electrons. Sov. Phys. Usp. 23, 227–244 (1980). [232] H. Tanaka, M. Isobe, and J. Yamamoto. Phys. Rev. Lett. 89, 168303 (2002). [233] S. Cotsakis, P. Leach, and G. Flessas. Phys. Rev. D 49, 6489 (1994). [234] F. C. MacKintosh and T. C. Lubensky. Phys. Rev. Lett. 67, 1169 (1991). [235] J. D. Noh and D. Kim. Phys. Rev. E 56, 355 (1997). [236] B. I. Lev. J. Mod. Phys. 7, 2366 (2016). [237] B. I. Lev. Mod. Phys. Lett. B 27, 1330020 (2013).

February 5, 2021

322

14:3

Applications of Field Theory Methods. . . - 9in x 6in

b4116-bib

Statistical Physics of Nonequilibrium Systems

[238] F. C. Adams, J. R. Bond, K. Freese, J. A. Frieman, and A. V. Olinto. Phys. Rev. D 47, 426 (1993). [239] C. W. Gardiner and P. Zoller. Quantum Noise. Springer, 2000. [240] D. F. Wells and G. J. Milburn. Quantum Optics. New Zealand, 2001; D. F. Wells and G. J. Milburn. Quantum Optics. Springer Science & Business Media (2nd ed.), 2008. [241] J. W. Gibbs. Elementary Principles in Statistical Mechanics: Developed with Special Reference to the Rational Foundation of Thermodynamics. Cambridge University Press, New York, 1902. [242] N. G. van Kampen. Stochastic Process in Physics and Chemistry. North-Holland, Amsterdam, 1990. [243] A. D. Linde. Phys. Lett. B 108, 389 (1982); 114, 431 (1982); 129, 117 (1983). [244] A. Albrecht and P. J. Steinhard. Phys. Rev. Lett. 48, 1220 (1982). [245] J. W. van Holten. Phys. Rev. Lett. 89, 201301 (2002). [246] P. Colet, F. De. Pasquele, and M. San Miguel. Phys. Rev. A 43, 5296 (1991).

page 322

February 5, 2021

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Index

absolute, 306 absorbed, 72 absorption, 284 accumulation, 265 acoustic phonon instabilities, 51 activation barrier, 174 additional assumptions, 81 additional field, 108 additional physical features, 49 additional restrictions, 276 adequate mathematical method, 2, 12 adiabatic thermodynamic, 269 admissible volumes, 197 after the transition, 289 all points of space, 93 all probable fluctuations, 108 all probable values, 93 alternative description, 270 a priori, 11 analogous, 306 analogously, 175 analytically derived, 308 anchoring energy, 226 antisymmetrize, 15 any degree of compression, 306 any governing parameter, 10 apply the functional integral, 1 appropriate explanation, 46 appropriate procedure, 5 approximate calculation method, 88 arbitrary coefficients, 301

arbitrary order, 5 point, 7, 85 quantity, 7 solution, 272 variable, 219 associated with finite values, 278 assumption of fluctuation, 10 astrophysical observations, 269 astrophysics, 10, 308 asymmetric behavior, 8 asymptotic value, 264 asymptotics, 91 attraction for bosons, 182 attraction range, 2 attraction-type gravitation energy, 32 auxiliary field variables, 141 auxiliary fields, 1 average energy, 177 average friction coefficient, 294 average kinetic energy 1, 178 average particle kinetic energy, 181 average values, 7, 259 averaging results, 6 avoid free-energy divergences, 5 background of the order parameter, 309 balance of gravitation and thermal energies, 306 barrier height, 227 323

page 323

February 5, 2021

324

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Statistical Physics of Nonequilibrium Systems

basic cases, 152 equation, 3 kinetic equation, 286, 311 solid-state property, 30 before relaxing towards thermodynamic equilibrium, 310 behavior of inhomogeneous states, 2 behavior of smaller-scale fluctuations, 6 behavioral specifics, 309 behavior of the system, 273 Bernoulli equation, 58 between the order parameter and its gradient, 55 Bjerrum length, 155 blisters, 51 Boltzmann activity, 180 equation, 271 distribution, 186 gas, 129 limiting case, 185 limit, 307 limiting case, 130, 176, 177 statistics, 3, 4, 128, 153, 167 Bose condensate, 3, 4 condensation, 127 gas, 4, 175 or Fermi, 15 particles, 201 statistics, 16 system, 180 Bose–Einstein condensation, 132, 180 both equilibrium and nonequilibrium cases, 12 both microscopic and phenomenological levels, 6 boundary condition, 119 Bragg, 41 Bragg–Williams free energy, 43 Bragg–Williams theory, 41 breaking it apart, 217, 223 brief encyclopedia, v

broken rotation symmetry, 70 Brownian particles may, 288 Brownian systems, 311 bubble formation, 296 bubble nucleation, 295 bubble size, 6 bubbles formed, 309 calculated exactly, 88, 104 calculation of path integrals, 102 cannot be solved completely, 167 cannot collapse, 197, 306 canonical equilibrium distribution function, 292 canonical description, 10 canonical moments, 85 canonical partition function, 284, 285 Cauchy formula, 125 cell, 229 cellular structure, viii, 216, 217, 310 cellular structures in colloids, ix certain idealized conditions, 283 chain-like structures, 216 change of packing, 32 changes the shapes, 37 character of interaction, 216, 217 characteristic dimensions, 9 charge carriers, 184 charge density, 155 chemical activity, 139, 160, 165, 172, 264, 276 chemical potential, 11, 143, 264, 277 chosen spatially inhomogeneous configuration, 48 classical and quantum systems, vii classical density-functional theory, 29 dynamical variable, 94 liquids, 30 manner, 123 objects, 13 path, 90 plasma system, 5 system, 94, 260 trajectories, 311

page 324

February 5, 2021

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Index close to equilibrium, 273 close-packing limiting case, 30 closely packed structure, 35 clusters, 152, 223 distribution, 3 formation, 3, 127 size for arbitrary temperatures, 179 sizes, 2 system, 3 formed, 305 coefficients, 48 collapse arises, 3 collapse in the system, 38 collective behavior, 7, 306 collective condensate behavior, 309 collective variables, 5 collisions, 74 colloidal fluids, 229 particle, 32 suspensions, 5 voids, 223 compact form, 139 compare the stability of the model condensate, 4 complete statistical description, 144 completely analogous, 78 completely determined, 283 completely determines, 277 completely reproduces, 109 complex plane, 305 complicated problem, 311 computer simulation, 222 concentration-dependence, 11 condensate, 4, 6 condensate collapses, 3 condensed phase, 206 “condensed” fluctuations, 9 condensate with the short-range attracting potential, 4 condensed matter, viii, 123, 216 condensed matter physics, 2, 5 condensed media, 65

page 325

325

condensed spatially periodic structures, 152 conditions for the formation, 2 configuration Hamiltonian, 140 configuration integral, 14 configurations, 92, 98 confine probable, 51 confine the spatially inhomogeneous state of the system, 49 confirmed experimentally, 139 confirms to the properties, 192 conservation law for the energy, 265 conservation laws, 259 conservation of the number of particles, 152, 308 conservation of the number of particles and energy, 261 consider the formation, 217 considerable fraction, 31 construction of the field theory, vii, 93 contain new phase bubbles, 306 continual integral, 311 continuous case, 146 continuum, 95 analogue, 146 case, 81 description, 226 limiting, 146 limiting case, 263 control parameters, 297 conventional numerical methods, 147 coordinate origin, 185 coordinate representation, 87 core, 119 coordinate space, 127 correct treatment, 309 correctly predicts, 73 correlation between, 287 correlation function, 84, 144 correlation length, 45, 101 correlation length diverges, 93 correlations, 299 cosmological model, 280 cosmological phenomenon, 297 cosmological problem, 311

February 5, 2021

326

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Statistical Physics of Nonequilibrium Systems

cosmological scenario, 56 cosmology reasoning, 280 Coulomb, 307 Coulomb interaction, 4, 152, 161 Coulomb law, 184 Coulomb screening, 37 Coulomb-like, 275, 308 Coulomb-like interaction, 4, 153 Coulomb-like system, viii, 4, 152, 155, 161 Coulomb-type, 307 Coulomb-type interaction, 187 coupling, 49, 55, 225 constant, 59 parameter, 161, 308 creates areas free of particles, 310 criterion, 228 critical fluctuations, 44 index, 44 point, 196 value, 3, 161 cross-section, 204 Cross and Newell, 51 crucial, 271, 297 crucial importance, 6 crucially differs, 210 crystal structures, 216 crystal-like ordering, 216 current, 184 cylindrical, 19 cylindrical molecule, 153 cylindrical sample, 308 d-dimensional, 11 d-dimensional hyperbolic lattice, 42 de Broglie wave, 16 Debye–Huckel theory, 67 decreasing amplitude, 280 defect, 49, 121 deformations, 51 degeneration condition, 272 degree of freedom, 45, 178 degradation processes, 284, 292 degree of violation enters, 296

delta function, 94 dense core, 31 density fluctuations, 269 matrix, 86, 87, 104 of states, 163 density-dependence, 12 density-functional theory, 30 description of many-particles systems, 1 detailed description of three-dimensional self-gravitating systems, 12 dielectric constant, 69 different conditions, 218 different nature, 65 different nature of interact, 229 different symmetries, 5 diffusion equation, 289 diffusion mass, 204 diffusion process, 76 dilute structures, 174 dipole moment, 7 discontinuous jump, 31 discrete Gaussian model, 113, 118 disordered, 33 configuration, 46 state, 31 dispersion, 8 rate, 174 dissipation, 293, 294 equation, 287 function, 293, 294 Langevin equation, 293 of energy, 312 relations, 286 displacements, 51 dissipative, 282 dissipative structures, 7 divergence, 307 diverse nature, 9 dominant contribution, 5 droplets immersed, 37 dual version of, 118 duality, 113

page 326

February 5, 2021

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Index duality transformation, 113 dust particle in a nonhomogeneous medium, 294 dusty plasmas, 152 dynamics, 288 dynamic aspects, 12 dynamic variables, 113 dynamical equation, 281 quantities, 7 theory, 281 viscosity coefficient, 282 variables, 7, 85, 259 dynamically changing Universe, 4 dynamics of cluster formation, 4 of cluster formation, 204 of the system, 282 of metastable states, vii, 73 earlier investigations, 2 effective, 77 entropy, 167, 170 free energy, 158 potential, 16, 227, 296, 309 screening length, 161 thermodynamic potential, 305 value, 228 eigenvalues, 15 eikonal equation, 52 elastic hard-sphere systems, 34 electromagnetic, 295 electrons on the helium surface, 158 ellipsoidal molecules, 19 emphasized that solution, 137 employ the equation of motion, 204 employ the hydrodynamical approach, 187 employ the quantum field theory approach, 1 employed the statistical theory of nonequilibrium processes, 2 energy barrier, 6 conservation, 266, 278

page 327

327

density, 261 dependence, 293 dissipation, 284 space, 311 entropy, 41, 120 environmental influence, 294 environmental system, 7 equal time scales, 307 equation of energy conservation, 172 equation governs, 273 equation of state, 147, 163, 174 equilibrium analogs, 8 and nonequilibrium statistical mechanics, v distribution, 310 ensemble vanishes, 288 sequence, 305 size, 307 size of the cluster, 3 solution, 286 states, 288, 291 systems, 123, 261 Euclidean expressions, 88 Euler–Lagrange equation, 47, 49, 52 evaluated exactly, 2, 5 evolution, 76, 281, 286 evolution of the system, 291 evolution of the Universe, 296 exact evaluation, 97 form, 15 result, 93, 308 solution, 56, 140 solutions for various spatial distributions, 153 exactly indisputable proof, 29 exactly reproduces, 137, 151 exactly solvable models, 139 exactly solvable models of statistical physics, viii exactly solvable three-dimensional model, 140 exciting problem, 5 exert a specific effect, 310

February 5, 2021

328

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Statistical Physics of Nonequilibrium Systems

exhibit collapse, 31 exhibit self-assembling into various structures, 4 exist far from thermal equilibrium, 283 existence of several finite-sized clusters, 173 existence of the Universe, 282 expand to infinity, 174 expanding the powers, 98 experimental data, 308 experimental observations, 8, 218 extended and applied, 308 external, 9 condition, 12, 260, 273 field, 6, 7, 45, 98, 206, 228 fluctuations, 296 noise, 7, 76 extrasolar giant planets, 269 extreme condition, 168, 264 extremum condition, 126 extreme nonequilibrium partition function, 12 extremum value of energy, 290 face-centered cubic, 31 factor of cubic structure, 38 false vacuum, 306 far from equilibrium, 11, 284 fast fluctuations, 309 favors the production, 120 Fermi and Bose particles, 3 energy, 181 function, 130 gas, 131, 175, 181 particles, 40 statistics, 16, 32 system, 132 wave vector, 39 ferromagnet, 50 ferromagnetic particle, 216 few macroscopic parameters, 284 few model systems, 2 Feynman path integral, 90

field “deformation”, 107 configuration, 108, 227 conjugate, 72 modification, 107 theory, 79, 295 variable, 86, 170, 265 field-dependence relief, 93 fifth order, 208 final equations of motion, 187 final state, 37 finite number of particles, 132 self-gravitating system, 183 spatial region, 265 volumes, 183 finite-size clusters, 2, 177 finite-size nucleus, 309 finite-size particles, 32 first-order phase transitions, vii, 5, 61, 73, 309 first approximation, 206 asymptotics, 185 integral, 149, 154, 158 fixed energy, 168 flow of nuclei, 75 fluctuating fluid membrane, 51 fluctuations, 309 fluctuation amplitude, 119 fluctuation dissipation, 311 fluctuations of arbitrary scales, 108 fluctuations of the medium, 65 fluid medium, 37 focal-conic defect structures, 50 Fokker–Planck equation, 75, 286, 290, 294, 300, 303, 311 Fokker–Planck-type equation, 76 force correlator, 75 formal expression, 96 formal integration, 262 formation conditions, 2 of giant planets, 269

page 328

February 5, 2021

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Index of spatially inhomogeneous particle and field distributions, 2 of structures, 1 time, 6 formed Universe, 282 formed in a pure elastic medium, 217 formed in equilibrium systems, 2 formulate the general approach, v Fourier series, 114 space, 33 transformation, 106 transforms, 109 Fourier-transformed, 113 fourth-order derivative, 51 fractional statistics of particles, 127 free-energy, 5, 41 density, 51 density functional, 60 functional, 49 freezing transition, 31 friction coefficient, 294 functional approach, 30 integral, viii, 1, 61, 98, 305 integral over occupation numbers, 61 integrals, v, 1 functionals associated with various sets, 48 fundamental physical background, 10 problem, 296 principles of thermodynamics, 283 scalar field, 47, 56, 280 scalar field in quantum field theory, 46 scalar field potential, 303 further applications, v galactic systems, 216 galaxy scattering, 280 gas bubble, 8

page 329

329

gas of interacting, 3 gas-dust matter, 308 Gauss integrals, 83, 106, 124, 141 Gauss Lattice models, viii Gauss model, 140 Gaussian, 108 distribution, 108, 301 integrals, 165, 262 integration, 99 law, 66 Gaussian-type integral, 93 general approach, 10 case, 73 description, 47 dissipation equation, 288 form, 308 formulas, 106 quantum case, 188 relation, 278 representation, 126 solution, 273 generating functional, 142 geometric representation, 7 geometrical presentations, 46 giant planets, stars, 308 Gibbs approach, 284 Gibbs ensemble, 260 Gibbs ensemble may provide a description of nonequilibrium stationary states, 274 Gibbs thermodynamic equilibrium, 11 Ginsburg–Landau equation, 281 global stability, 297 Gordon equation, 308 governed by long-range interaction, 10 governing parameter, 7, 8, 76, 309 governing parameter fluctuations, 8 gradient term, 45 gradient theory, 47, 50 gradient theory of phase transition, 50 grand canonical ensemble, 142 gravity factor, 273

February 5, 2021

330

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Statistical Physics of Nonequilibrium Systems

gravothermal catastrophe, viii, 189, 192 gravitating Bose gas, 199 gas, 186, 307 gas model, v, viii, 148 system, 222 gravitation energy, 307 force, 182 instability, 269 systems, 11, 310 gravitational attraction, 126 collapse, 10 formation of stars, 11 potential, 11 Green function, 120, 128, 146, 262 Gross–Pitaevsky equation, 3, 4, 206 ground state, 46, 303 H-theorem, 286 Hamiltonian, 13, 23, 261 Hamiltonian of the system, 85 hard core and finite-range attraction, 34 hard interactions, 30 hard sphere freezing, 31 hard spheres, 30, 199, 201 hard spheres model, viii, 2, 134, 137, 148 hard-sphere potentials, 30 harmonic fluctuations, 99 lattice model, viii, 110 potential of the trap, 206 heat capacity, 14 height of the barrier, 208 Heisenberg model, 27 Heisenberg system, 120 helpful in testing, 4 hexagonal closely packed, 31 hexagonal structure, 38 high-temperature limiting case, 3 highly efficient approach, 93

highly sensitive, 310 homogeneity, 297 homogeneous and inhomogeneous particle distributions, 2 homogeneous or inhomogeneous, 305 homogeneous system, 51 Hubbard–Stratonovich representation, 308 Hubbard–Stratonovich transformation, v, viii, 96 Hubble parameter, 298 hump type, 186 hydrodynamic approach, 307 hydrodynamic calculations, 269 hydrodynamical approach, 187 ideal classical, 128 ideal classical and quantum gases, viii ideal conditions, 283 identically equal, 90 important problems, v inequality, 193 in power series, 150 in quantum gases, 182 in terms of the averaged order parameter, 6 in the three-dimensional space, 155 incorporate physical information, 30 independent factor, 81 independent fluctuations, 108 indeterminate result, 120 indicate the formation of a crystal-like structure, 162 individual particle states, 123 individual points, 109 infinite system, 189 infinitely high kinetic energy, 185 influence of random changes, 294 of the environment, 312 of the external factors, 3 of fluctuations, 78 of the environment, 283 information on time scales, 174

page 330

February 5, 2021

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Index inhomogeneity of self-gravitating systems, viii inhomogeneous distribution, 260 distribution in systems of particles, viii particle distributions, 148, 311 solution, 265 stable structures, 51 initial concentration, 217 cubic cell, 36 homogeneous state, 9 time, 291 instability, 4, 310 instability threshold, 11 integral asymptotics, 89 integral transformations, 5 intensively studied, 1 interacting Bose particles, 4 defects, 113 many-particle systems, 1 particle systems, vii, 13 particles, 2, 3, 93, 265, 310 interaction between particles, 218 interaction intensity, 9 interaction type, 3 intermediate variables, 87 internal field fluctuations, 296 interparticle distance, 226 introduce an additional hypothesis, 12 invariants, 51 inverse matrix, 124 inverse operator, 128, 174, 306 inverse of the interaction matrix, 165 involve spatially inhomogeneous field distributions, 278 involving higher orders, 49 Ising model, 20, 41, 96, 164 isolated systems, 282 isotherms, 196 isotropic phase, 72

page 331

331

isotropy, 297 Ito form, 289 jump discontinuity, 45 Kasterliz and Thouless, 120 kinetic dynamics, 174 energy, 15, 132 properties, 294 theory, 310 known model systems, 2 known parameters, 171 knowledge, 280 known expression, 151 laboratory conditions, 4 Lagrange equation, 176 Lagrange multipliers, 260 Landau free energy, 227 mean-field theory, 43 model of phase transitions, 48 theory, 43, 46, 73 Lagrangian, 297 Langevin equation, 75, 288, 294, 300 Laplace operator, 146, 175, 263, 270 Laplace pressure, 221 Laplacian pressure, 37 large particle in a suspension, 294 latent heat, 72 lattice constant, 95 Coulomb gas, 113, 118 Green function, 118 model, 95, 96 sites, 123 layer thickness, 212 length of the statistical instability, 273 Lennard–Jones interaction, 18 lifetime in the metastable state, 298 lifetimes diverge, 11 likely charged particles, 215 limiting case, 95

February 5, 2021

332

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Statistical Physics of Nonequilibrium Systems

linearized equation of motion, 187 linear harmonic oscillator, 103 linear solutions, 53 Lipschitz solutions, 52 liquid crystal, 50, 70, 216 liquid crystal colloids, 216, 217 liquid crystals, dusty plasmas, low-dimensional structures in solids, 1 liquid crystals, superconductors, 65 liquid phases, 13 liquid–solid transitions, 29 local change of thermodynamic parameters, 11 changes, 76 conservation law, 289 entropy, 267 entropy landscape, 282 equation of state, 278 equilibrium, 73, 260 equilibrium distribution, 260, 274 equilibrium ensemble, 260 equilibrium ensemble exactly, 274 equilibrium stationary distribution, 260 maxima of the thermodynamic potential, 260 order parameter, 43 theory, 109 variations, 10 localization within a limited space, 127 locally isothermal, 269 long range action, 206 and short-range interactions, 147 attracting potential, 4 attraction, 2, 4 attraction and short-range repulsion, 32 behavior, 11 gravitation inter, 270

nature of the Coulomb interaction, 157 order altogether, 119 long-wavelength approximation, 101 long-wavelength expansion, 36, 226 low temperatures, 133 low-temperature limiting case, 92 low-temperature states, 91 lowest local density, 217 macrobubbles, 9, 309 macroscopic order parameter, 9 macroscopic system, 283, 284, 286, 311 magnetization vector in theory of magnetism, 46 many cells, 94 many traditional problems, 1 many-body systems, 10 many-dimensional case, 89 many-particle, v interactions, 104 problem, 126 systems, 4, 217 mass coefficient, 109 mathematical calculations, 7 mathematical complications, 179 mathematical point, 50 mathematically equivalent, 300 mean time required, 76 mean-field approximation, 278, 279 form, 45 order parameter, 99 theory, v, viii, 10, 93, 95, 98 theory consists, 98 mechanism that limits the energy, 294 mechanisms, 306 medium fluctuations, 76 melting-like phenomena, 5 metastable particle distributions, v spatially inhomogeneous states, 1 states, v, 174

page 332

February 5, 2021

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Index methods, v methods of quantum field theory, 305 microcanonical description, viii description of gravitating systems, 163 distribution, 291 distribution function, 163 ensemble, 10, 163, 173 partition function, 165, 167, 169 micromagnetics, 50 microscopic and phenomenological levels, 9 microscopic treatment, 6 minimum of the free energy, 306 model formally are meaningful, 109 model of hard solution of substitution, 24 model of hard spheres, vii, 29 model of phase transitions with coupling, 56 model systems, viii, 123 models of statistical physics, vii, 20 models with attraction and repulsion, viii, 209 modifications, 284 molecules, 70 momentum, 14 momentum space, 45 monotonously, 307 morphological instabilities, 282 multisoliton solutions, 305 mutually dependent, 278 n-particle correlation, 109 n-vector model, viii, 112, 113 natural condition, 272 natural general extension, 261 natural inflation, 280 nature and intensity of interaction, 166 nature of the phenomenon, 197 necessary condition, 282, 303 necessary thermodynamic conditions, 265

page 333

333

negative scattering, 4 negative scattering lengths, 3 negative specific heat, 10 nematic phase of liquid crystals, 70 new approach, 1 degrees of freedom of motion, 119 energy state, 312 “equilibrium” state, 290 phase bubbles, 6 spatial structures, 3 stationary state, 284 substances, 1 variables, 104 Newtonian attraction, 174, 184 gravitation interaction, 275 interaction, 146, 222 potential, 128 no divergences, 308 no singularities, 174 noise-induced formation, 297 phase transition, 8, 9 transition theory, 76 transitions, 76 nonequilibrium, 11, 310 conditions, 282 distribution, 311 distribution function, 286, 289 dynamics, 299 dynamics of universe, ix fluctuations, 301 gravitating systems, ix, 259 medium, 8 quasi-stationary states, 282 state, 259, 260 stationary states, 259 statistical description, 310 statistical operator, 11, 260, 261, 263 statistical operator approach, 11 system, ix, 7, 259, 260, 282, 288 nonequivalence, 6

February 5, 2021

334

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Statistical Physics of Nonequilibrium Systems

nonequivalent minimum, 6, 8 nonextensive parameter, 269 noninteracting particles, 167, 218 nonideal gas, 38 nonlinear differential equation, 207 equation, 3, 149, 308 integral-differential equation, 207 Langevin equation, 289, 299 model, 120 particle-velocity dependence, 294 solutions, 54 nonlinearity factor, 109 nontrivial behavior, 10 nonuniform hard-sphere systems, 30 nonzero order parameter, 5 normal uncollapsed states, 31 normalization condition, 170, 172, 264, 268, 277, 285 nucleation of voids, 223 nucleus, 76 number of fermions, 39 number of particles, 155, 168, 183 number of particles grows, 11 observation temperature, 8 observed experimentally, 46 observed parameters, 9 observed symmetry breaking, 31 obvious from the definition, 142 occupation numbers, 7, 123, 164 one-dimensional case, 89, 153 one-loop expression, 100 open systems, 1 optimum states, 2 order-parameter amplitude, 119 order-parameter gradient, 47 ordered structures, 5 ordinary Brownian particles, 283, 288, 293 ordinary entropy, 169, 174 original density-functional theory, 30 orthonormal system, 15 over the auxiliary fields, 305

packet diffusion, 4 particle comp, 226 density, 267 distribution inhomogeneity, 11 distributions, v solution is disordered, 226 particles immersed in a liquid crystal, 216, 310 particular direction specified, 70 partition function, 1, 2, 86, 94, 137, 308 partition function are dominant, 126, 305 partition function computation, 105 path integrals, viii path integration, v, 79 path-integration method, 79 Pauli principle, 181 peculiar phenomena, 5 penetrate the coordinate origin, 185 percolation transition, 229 perfect media, 5 periodic cosine potential, 118 distribution of, 308 function, 118 structure, 155 period of the structure, 155 perturbation theory, 1 perturbation term, 100 phase space, 284 phase transitions, v phase transitions accompanied, 1 phase transitions in the two-dimensional case, 122 phenomenological approach, 9 continuum gradient field theory, 96 description, 48 expansion, 93 expression, 10 free-energy, 48 modeling, 51

page 334

February 5, 2021

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Index models, 46 thermodynamics, 259, 274 treatment, 6 physical characteristics, 7 conditions, 171 interpretation, 289 justification, 51 manifestation, 2 meaning, 87 parameters, 259 reactions, 273 reasons, 36 understanding, 2 plasma, 184 plasma frequency, 307 plasmas, colloidal particles, electrolyte solutions, electron gas in solids, 4 Poisson formula, 118 Poisson summation formula, 118 polaron, 186, 307 polytrophic dependence, 265 possible in the system of attracting particles, 3 postulate, 59 potential barrier, 134, 296 fluctuations, 296 possess, 56 Potts model, viii, 113 Potts model partition function, 114 power series, 95 power series expansion, 36, 95 power-law velocity distributions, 269 pressure, 133, 163 pressure in the packing, 34 pressure, chemical potential, density, 278 pressure, compressibility, 147 primarily regularizing, 11 principle of integration, 108 probability description, 10 “probability” potential, 77 probable

page 335

335

approach, 295 changes, 7 configurations, 98 definition, 284 deviation, 7, 285 distributions, 259 fluctuation configurations, 108 fluctuations, 6, 8 formation of structures, 160 interactions, 81 motion, 312 paths, 86 perturbations, 49 processes, 300 spatial states, 262 stationary states, 312 structures, 310 trajectories, 88 process of collapse, 306 processes of energy exchange, 283 proliferation, 119 proper dispersion, 9 properties of the medium, 310 provides a possibility, 3 provokes a collapse, 36 pure Coulomb repulsion, 126 purely attractive gravitation interaction, 31 purely Coulomb system, 308 purely repulsive interaction, 274 quadratic dependence, 93, 308 quadratic field-dependence, 104 quadratic form, 88 quadratic term, 89 quantitatively predict, 11, 310, 311 quantum case, 272 contribution, 311 energy, 182 energy compresses, 182 field theory, v, 276 gases, 128 mechanics, 88 particles, 16

February 5, 2021

336

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Statistical Physics of Nonequilibrium Systems

physics, 272 pressure, 3, 207 systems, 79, 86 quantum-field-theory approach to, 152 quantum-mechanics operator, 94 quasi-classical approximation, 88 quasi-equilibrium state, 174 quasi-equilibrium systems, v quasi-stationary, 283 random fluctuations, 108 random force correlator, 75 random influence, 299 rapidly quenching, 56 real period of the structure, 159 real space, 107 realistic distributions, 273 realistic Ising model, 139 realized under the thermodynamic condition, 311 relation between the dissipation in the system and the diffusion in the stationary case, 290 reasoning is analogous, 188 reciprocal dimension, 204 reciprocal temperature, 14, 91, 163 referred informally, 51 regarding formulating, 30 regions free from particles, 217 relations for nonequilibrium systems, 311 relatively long time, 174 relaxation time, 260, 307 relevant conditions, 311 derivative, 52 energy, 300 equations for, 144 Lagrange equation, 185 minimum, 306 order parameter, 48 physical process, 289 statistical ensembles, 259 systems, 148

representation, 1, 7, 305, 312 repulsing particles, 307 repulsion for fermions, 182 repulsion range, 2 repulsive Fermi, 307 require further studies, 1 required behavior, 6, 9 required thermodynamic, 260 residual interaction, 3, 305 restrict the analysis, 7 reversible and irreversible clustering, 223 saddle point, v, 132, 264 approximation, 152, 160, 308 configuration, 309 equation, 149, 154, 171, 209, 265, 276 method, vii, 1, 92 solution, 99, 170 state, 126, 160 saddle point method, 264 saddle states, 282 saddle states of nonequilibrium systems, ix Saslaw approach, 218 satisfy the extremum condition, 305 scalar field, 295 scalar field functions, 7 scalar order parameter, 51 scalar, vector, or tensor of arbitrary rank, 7 scales of fluctuations, 109 scattering length, 4 scattering of matter, 280 Schr¨ odinger equation, 85 screened Coulomb repulsion and attraction, 209 screening Coulomb, 128, 146 screening length, 146 second derivative, 89 second order, 207 second-order phase transitions, vii, 41, 93

page 336

February 5, 2021

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Index second-rank tensor in liquid-crystal theory, 46 select the states, 1, 143 selection is combined, 309 selection of probable states, 6 selection of states, viii, 9, 216, 309 self-assembly, 5 self-consistent approximation, 113 self-consistent field, 225 self-consistent field approximation, 100 self-consistently, 68 self-gravitating, 10 gas collapses, 184 system, 10, 11, 167, 173, 174, 259, 261 semi-phenomenological field theory, 93 separate macrosystem, 283 separate the system, 2 separated order parameter, 8 short-range attraction, 137 behavior, 31 cutoff, 31 forces, 139 repulsion, 2 short-wavelength distortion, 95 shown that cluster formation, 3 similar character, 260 master equations, 188 nature, 217 systems, 218 simple examples, 48 diffusion equation, 291 heuristic argument, 120 models, 93 simplest Ising model, 306 simplification, 51 simplified version, 57 sine-, 308 sine-Gordon equation, 154 sine-Gordon model, 126

page 337

337

single vortex, 120 singular perturbation models, 49 slower fluctuations, 309 slowest subsystem, 6, 309 small perturbations, 5 small ratio, 16 small volumes, 133 smectic-A liquid crystal, 50 soap froths, 223 soft body, 310 soft cores, 37 soft-matter systems, 4 solar nebula, 269 solid state, 38 soliton solution, 3, 153, 177, 193 solution with homogeneous distribution, 49 solved in terms of the proposed approach, 3 solvent induced phase separation, 229 solving relevant thermodynamic relations, 12 solving the problem under consideration, 4 some approximate methods, 105 some special interparticle interaction, 6 some specific properties, 56 space-coordinate dependence, 272 space-dependent, 56 space point, 265, 278 spatial boundary conditions, 155 dependence, 278 derivative, 49, 51 distribution function, viii, 174, 190 distributions, 60 inhomogeneity, 4 inhomogeneous particle distributions, 273 spatially homogeneous to inhomogeneous states, 55 inhomogeneous, 216, 265

February 5, 2021

338

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Statistical Physics of Nonequilibrium Systems

inhomogeneous distribution of particles, 177, 228 inhomogeneous distributions, 305 inhomogeneous equilibrium, v inhomogeneous particle distributions, 123 inhomogeneous phases, v nonuniform distribution, 108, 310 periodic distributions, 162 special attention, 9 special cases, viii, 102 special emphasis, 10 special interaction, 18 specialized explanations, 217 specific dynamical variables, 259 examples, 47 features of the matter, 20 heat, 45 methods, 5 physical system, 7 properties, 7 systems, 7 specifics of the system, 283 spherical coordinate system, 63 spherical isolated stellar system, 271 spin, 7 spinodal decomposition, 56 spontaneous symmetry breaking, 11, 295, 310 stabilization of regions, 217 stable, 307 solutions, 11 state, 46 states, 260 thermodynamic behavior, 309 standard and gradient theories, 55, 60 cosmological models, 295 dimensionless free energy, 48 expression, 91 Jeans length, 269

methods, 4 model, 47 starting from the nonequilibrium statistical operator, 274 state under consideration, 3 state-selection methods, 10 states associated, 2 stationary behavior, 296 external conditions, 260 macroscopic behavior, 10, 78 nonequilibrium distribution function, 290 point, 89 solution, 286, 290, 292, 311 states, 1 stationary-phase method, v, vii, 88 statistical, 10 analysis, 9 description, v, 1, 3 field theory, 5 interaction, 14 mechanics, 4, 86, 206 motivation, 280 operator, 260 physics, v potential, 182 theory, 305 thermodynamics, 259 steady states, 311 Stratonovich form, 289, 300 Stratonovich presentation, 289 strong interaction, 157 strongly constrained models, 96 structural factor, 38 parameters of the system, 85 transformations, 8 transition, 38 structures with lower ordering, 217 sufficiently larger number, 119 suitable solution, 272 superfluid motion, 119 superheated liquid, crystallization in a supercooled liquid, or

page 338

February 5, 2021

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Index superconductor in a magnetic field, 8 surface energy, 212, 228 surfactant solutions, colloids in various solvents, and dust particles in plasmas, 4 susceptibility, 72, 73, 144 symbolic form, 87 symmetry of particle wave functions, 182 symmetry violation calculated, 8 systems with repulsive interaction, ix tailored to the particles with gravitational interaction, 311 temperature, volume, 7 temperature-dependence, 60 tendency to disperse, 174 tends towards a nonequilibrium state, 174 tensor, 70 thermal bath, 283 thermodynamic characteristics, 126, 152 conditions, 192, 307 ensembles, 10 equilibrium, 310 functions, 259 limit, 3 limiting, 5 limiting case, 127 parameter, 7, 140, 259, 260, 273, 278 potential, 8 potential nonlinearity, 8 properties, 221 quantities, 7, 85 reason, 173 relation, 260, 264, 267, 278 tools, 11 treatment, 218 kinetic, electronic, and electromagnetic properties, 1 thermodynamical functions, 133 thermodynamically stable, 5, 123, 166

page 339

339

thermodynamically stable particle distributions, 1, 126, 305 thermodynamically unstable, 11 thermostat, 7 thermal de Broglie wavelength, 276 energy, 16 length, 16 thin films, 51 third-order invariant, 73 three dimensions, 148 three-dimensional case, 103, 159, 160, 261, 270 models, 139 space, 92 structure, 159 systems, 282 lattice system, 147 time dimensions, 87 time of relaxation, 6 theoretical physics, 1 theoretical treatment, 5 theory of phase transitions, v time evolution, 298 topological charge, 121 two-dimensional phase transition, 223 two-state Potts model, 114 topological point defects, 120 topologically stable defect, 119 total magnetic moment, 41 total potential, 68 towards a thermodynamically stable state, 6 trace operation, 93 traditional method, 130 transformation Jacobian, 83 transformation of the length, 107 transforms, 75 transient dynamics, 297 transitions between crystalline phases, 5 trapped in quasi-stationary states, 11 trivial manner, 34 tunneling probability, 296 tunneling through, 296

February 5, 2021

340

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Statistical Physics of Nonequilibrium Systems

tunneling through the barrier, 3 two auxiliary fields, 306 two-dimensional, 120 case, 155 Coulomb gas, 121 Coulomb-gas models, viii, 119 Coulomb-like system, 157 electron liquids, 155 lattice, 22 model, 113 phase transition, 223 square lattice, 113 system, 155 two homogeneously charged lines, 155 two model systems with opposite interaction mechanisms, 140 two-phase coexistence, 56 two-state Potts model, 114 type of interaction, 2 typical physical situation, 2 unambiguously related, 278, 305 uncertainty, 55 uncertainty principle, 207 uncompensated after the cluster formation, 3 understanding condensed, 13 undetermined Lagrange multiplier, 193 unification, 60 unification of the theories of phase transitions, vii, 46 unified approach, v, 123 unified manner, 140 uniform spatial distribution, 119 unit vector, 50, 70 universal sequence, 5 universality principles, 139 Universe, 7, 299 unknown reciprocal temperature, 171 unperturbed collisionless gas, 269 unstable environment, 296 usual entropy, 172

usual colloid, 216, 217, 225 usually surrounded, 7 vacuum, 297, 299 valuable for applications, 2 Van-der-Waals type equation, 34 van der Waals attractive interactions, 223 Van-der-Waals equation, 18 vanish on the boundaries, 119 vanishes, 90 vanishing coupling constant, 60 variable coefficients, 207 variation, 76, 260 variation scale, 309 various cosmological models, 295 crystal structures, 5 crystals, 51 distributions of particles, 265 fields, 5 methods, v physical situations, 216 spatial processes, 312 statistics, 3 structures, 224 systems, 218 types, 218 vector quantity, 50 verified experimentally, 307 very large screening length, 11 Villain, viii and Gauss Lattice models, 116 model, 118 version of the xy-model, 118 violate the equilibrium condition, 37 virial coefficient, 137, 222 void formation, 221 voids without particles, 218, 229 volume grows infinite, 5 volume of the sample, 119 volume of the system, 69 volume-averaged gradient, 119 vortex density correlation function, 121

page 340

February 5, 2021

15:25

Applications of Field Theory Methods. . . - 9in x 6in

b4116-index

Index vortex part, 121 vortices, 119, 121 wave, 3 wave function, 15, 206 wavelength shorter, 6 weak external effects, 310 weak interaction between particles, 221 weakly ionized plasma, 159 weakly nonideal, 179 weight assigned, 94 weights, 305 well-defined methods, 284 well-known approach, 224 equilibrium distributions, 299

page 341

341

formula, 14 result, 12 standard model, 48 wetting effect, 227 wetting solvent phase, 229 white noise, 300 Wigner crystal, 152, 161, 308 Williams, 41 xy-model, 118 zero-order modified Bessel function, 298 zero-order phase transition, 31